Understanding the efficiency of ionic liquids-DMSO as solvents for carbohydrates: use of solvatochromic- And related physicochemical properties

Article abstract of DOI:10.1039/d0nj02258f

The physical dissolution of carbohydrates (cellulose, chitin, and starch),i.e., without the formation of covalent bonds requires the solvent to possess certain physicochemical properties. Concentrating on cellulose, the solvent should act both as a Lewis acid and a Lewis base, and disrupt the present hydrophobic interactions, as the biopolymer exhibits amphiphilic characteristics. The quantification of the relative importance of these physicochemical properties helps in predicting the solvent structures, which are expected to be efficient as cellulose solvents. Ionic liquids (ILs) are extensively used as carbohydrate solvents because they disrupt the intramolecular-, intermolecular-, and hydrophobic interactions within the biopolymer structure, leading to its dissolution. Solvatochromic substances (probes) are especially sensitive to one or more of the above-mentioned biopolymer-solvent interactions. Consequently, they are used to predict and rationalize the solvent efficiency. The solvent parameters (descriptors) most widely employed are empirical polarity,E_T(probe), Lewis acidity (SA); Lewis basicity (SB), dipolarity (SD), and polarizability (SP); S refers to the solvent. We synthesized 18 ILs, including derivatives of imidazole, 1,8-diazabicyclo[5.4.0]undec-7-ene, and tetramethylguanidine; the corresponding anions are carboxylates, chloride and dimethylphosphate. We used solvatochromic probes to calculate the descriptors of IL-DMSO (at fixed DMSO mole fraction of 0.6; 40 °C), and correlatedE_T(probe) with the other descriptors. We also tested the correlations by using a molar volume of the IL (V_M) instead of SD, and the Lorentz-Lorenz refractive index functionf(n) of the IL-DMSO mixture instead of SP. The quality of the regression analysis increased noticeably when we limited the ILs correlated with those based on imidazole (13 ILs), and used (V_M) andf(n). The regression coefficients showed that SA is the most important descriptor; the solvent empirical polarity is inversely dependent onV_M. The value off(n) shows the importance of hydrophobic interactions. By using different probes, we showed that the observed small contribution of SB reflects the steric crowding around the positive nitrogen atoms in some probes. The results obtained help in selecting ILs as solvents for cellulose and other carbohydrates, based on the expected strength of their interactions with the biopolymers. Therefore, using solvatochromism for solvent efficiency screening saves labor and cost.

Full text of DOI:10.1039/d0nj02258f

View Article Online
View Journal
NJC
New Journal of Chemistry
Accepted Manuscript
A journal for new directions in chemistry
This article can be cited before page numbers have been issued, to do this please use: O. El Seoud, T. D.
A. Bioni, M. T. Dignani and M. L. de Oliveira, New J. Chem., 2020, DOI: 10.1039/D0NJ02258F.
Volume 42
Number
March 2018
Pages 3963-4776
This is an Accepted Manuscript, which has been through the
Royal Society of Chemistry peer review process and has been
accepted for publication.
6
21
NJC
New Journal of Chemistry
rsc.li/njc
A journal for new directions in chemistry
Accepted Manuscripts are published online shortly after acceptance,
before technical editing, formatting and proof reading. Using this free
service, authors can make their results available to the community, in
citable form, before we publish the edited article. We will replace this
Accepted Manuscript with the edited and formatted Advance Article as
soon as it is available.
You can find more information about Accepted Manuscripts in the
Information for Authors.
Please note that technical editing may introduce minor changes to the
text and/or graphics, which may alter content. The journal’s standard
Terms & Conditions and the Ethical guidelines still apply. In no event
shall the Royal Society of Chemistry be held responsible for any errors
or omissions in this Accepted Manuscript or any consequences arising
from the use of any information it contains.
ISSN 1144-2546
PAPER
Chechia Hu, Tzu-Jen Lin et al.
Yellowish and blue luminescent graphene oxide quantum dots
prepared via a microwave-assisted hydrothermal route using
H2O2 and KMnO4 as oxidizing agents
rsc.li/njc

Page 1 of 25
New Journal of Chemistry
[
1]
1
2
3
4
5
6
7
8
9
View Article Online
DOI: 10.1039/D0NJ02258F
Understanding the efficiency of ionic liquids-DMSO as solvents for
carbohydrates: Use of solvatochromic- and related physicochemical
properties
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Thaís A. Bioni, Mayara L. de Oliveira, Marcella T. Dignani, Omar A. El Seoud,*
Institute of Chemistry, the University of São Paulo, 748 Prof. Lineu Prestes Av.,
0
5508-000 São Paulo, SP, Brazil. Email (elseoud.usp@gmail.com)
Abstract
The physical dissolution of carbohydrates (cellulose, chitin, and starch), i.e.,
without formation of covalent bonds requires that the solvent possess certain
physicochemical properties. Concentrating on cellulose, the solvent should act both as
Lewis acid, and Lewis base, and disrupt the present hydrophobic interactions, as the
biopolymer has amphiphilic character. Quantification of the relative importance of
these physicochemical properties helps in predicting solvent structures expected to be
efficient as cellulose solvents. Ionic liquids (ILs) are extensively used as carbohydrate
solvents because they disrupt intramolecular-, intermolecular-, and hydrophobic
interactions within the biopolymer structure, leading to its dissolution. Solvatochromic
substances (probes) are especially sensitive to one, or more of the above-mentioned
biopolymer-solvent interactions. Consequently, they are used to predict, and
rationalize solvent efficiency. The solvent parameters (descriptors) most widely
employed are: empirical polarity, E (probe), Lewis acidity (SA); Lewis basicity (SB),
T
dipolarity (SD), and polarizability (SP); S refers to solvent. We synthesized 18 ILs,
including derivatives of imidazole, 1,8-diazabicyclo[5.4.0]undec-7-ene, and
tetramethylguanidine; the corresponding anions are carboxylates, chloride and
dimethylphosphate. We used solvatochromic probes to calculate the descriptors of IL-
DMSO (at fixed DMSO mole fraction of 0.6; 40 C), and correlated E (probe) with the
T
other descriptors. We also tested the correlations by using the molar volume of the IL
(
VM) instead of SD, and Lorentz-Lorenz refractive index function (n) of the IL-DMSO
mixture instead of SP. The quality of the regression analysis increased noticeably when
we limited the ILs correlated to those based on imidazole (13 ILs), and used (V ) and
M

New Journal of Chemistry
Page 2 of 25
[
2]
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
View Article Online

(n). The regression coefficients showed that SA is the most important descriptor;
DOI: 10.1039/D0NJ02258F
solvent empirical polarity is inversely dependent on V . The value of (n) shows the
M
importance of hydrophobic interactions. By using different probes, we showed that
the observed small contribution of SB reflects steric crowing around the positive
nitrogen atom in some probes. The results obtained help in selecting ILs as solvents for
cellulose and other carbohydrates, based on the expected strength of their
interactions with the biopolymers. Therefore, using solvatochromism for solvent
efficiency screening saves labor and cost.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Key Words
Solvatochromism, solvatochromic descriptors, ionic liquids, cellulose dissolution,
correlation of solvent descriptors, biopolymer-solvent interactions.
1- Introduction
Several solvents and their mixtures dissolve carbohydrate polymers, e.g., cellulose,
chitin and starch physically, i.e., without covalent bond formation.1-9 The resulting
solutions are used to regenerate the biopolymer, e.g., as fibers, films, and
nanoparticles, and to synthesize derivatives, e.g., esters and mixed esters.10-14
Considering cellulose, derivatization under homogeneous reaction conditions leads to
products with better control of regioselectivity and degree of substitution both in the
biopolymer structural unit (anhydroglucose unit; AGU), and along its backbone.
Consequently, the products thus obtained show reproducible properties, e.g.,
dissolution in solvents and their mixtures.14,15 As cotton production is not expected to
16
meet world demand for fabrics, the use of cellulose from other sources, in particular
wood is expected to expand. Use of the latter involves biopolymer dissolution followed
by fiber regeneration in a suitable bath. Consequently, there is intense and sustained
interest in the physical dissolution of cellulose.17
For the above-mentioned biopolymers there is interest is understanding the
following aspects: (i) the mechanism of their physical dissolution; (ii) assessment of the
solvent physicochemical properties that lead to efficient biopolymer dissolution. As
both issues were studied in much more details for cellulose, we focus on the
dissolution of this biopolymer, although the conclusions apply to other biopolymers, at

Page 3 of 25
New Journal of Chemistry
[
3]
1
2
3
4
5
6
7
8
9
View Article Online
DOI: 10.1039/D0NJ02258F
least qualitatively. Regarding (i), there is agreement that the solvent should disrupt the
inter- and intramolecular hydrogen-bonding (H-bonding) present, as well as the
hydrophobic interactions because cellulose has amphiphilic character.18,19 This raises
the following question: Is it possible to quantify the relative importance to cellulose
dissolution of H-bonding and hydrophobic interactions? In the present work we discuss
a strategy to answer this question with emphasis on the use of solvatochromic
parameters (vide infra the definition of solvatochromism) and other, alternative
physicochemical properties of the solvent.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
We synthesized 18 ionic liquids (ILs) that are known (based on literature) to
dissolve cellulose.4,5,9,14,20-23 These include derivatives of imidazole; 1,8-
diazabicyclo[5.4.0]undec-7-ene (DBU), and tetramethylguanidine (TMG). Of these,
fifteen IL have carboxylate anions, and three have chloride and dimethylphosphate
24
anions. It is known that binary mixtures of ILs and molecular solvents (MSs) are, in
many cases, more efficient as cellulose solvents than pure ILs.18,25 Consequently, we
used these ILs as solutions in dimethyl sulfoxide (DMSO), at a fixed mole fraction of the
MS; DMSO = 0.6. We employed this binary solvent composition because our previous
data on cellulose dissolution in mixtures of several ILs and DMSO showed biopolymer
.23
dissolution maxima at DMSO of ca. 0.6 We calculated the solvatochromic parameters
of these binary mixtures (BMs), and correlated them. We also examined the use of
other physicochemical properties of the solvent, namely IL molar volume (V ) and the
M
Lorentz-Lorenz refractive index function (n) of the BM. Our data showed that limiting
the molecular structure of the ILs to imidazole derivatives (13 ILs) resulted in
satisfactory multi-parameter correlation whose quality was further enhanced by
including (V ) and (n) instead of (solvatochromic-based) solvent dipolarity and
M
polarizability, respectively. The most relevant solvent parameters that contribute to its
empirical (or total) polarity, hence to biopolymer dissolution, are those related to H-
bonding.
2
-Experimental
We purchased all reagents and solvents from Sigma-Aldrich or Synth (São
Paulo) and purified them as given elsewhere.26

New Journal of Chemistry
Page 4 of 25
[
4]
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
View Article Online
DOI: 10.1039/D0NJ02258F
2
.1- Synthesis of the ionic liquids
Table 1 shows the molecular structures of the synthesized ILs and their
physicochemical parameters of interest. We discuss later the calculation of these
1
1
parameters. We list the H NMR data (Varian Innova-300, 300 MHz for H, CDCl ) for all
3
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ILs synthesized in Table ESI-1 (Table 1 of Electronic Supplementary Information).

Page 5 of 25
New Journal of Chemistry
[
5]
1
2
3
4
5
6
7
8
9
Table 1: Molecular structures of the ionic liquids (ILs) employed; physicochemical properties of the binary mixtures IL-DMSO, at DMSO = 0.6,
o
a
and 40 C, and molecular volumes of the pure ILs.
ET(WB);
V ;
M
Cation
Anion
SA
SB
SD
SP
n
f(n)
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
3
kcal/mol
57.72
56.97
58.39
57.70
57.40
57.04
56.08
55.24
56.40
55.70
56.78
56.25
55.73
cm /mol
CH3
N
Acetate
Benzoate
0.156
0.157
0.170
0.158
0.159
0.155
0.135
0.136
0.151
0.147
0.149
0.150
0.134
0.928
0.791
0.716
0.738
0.720
0.935
0.958
0.799
0.731
0.764
0.742
0.744
0.965
1.169
0.808
0.982
1.196
1.019
0.927
1.211
1.253
1.116
1.219
1.097
1.109
1.042
0.865
0.919
0.843
0.876
0.863
0.871
0.873
0.937
0.848
0.831
0.86
183.03
237.88
153.82
166.47
208.82
199.84
199.59
254.44
170.38
243.49
183.03
225.38
216.40
1.4988
1.5189
1.4990
1.5045
1.5030
1.5023
1.5057
1.5248
1.5031
1.4938
1.5022
1.5033
1.5074
0.2959
0.3024
0.2932
0.2972
0.2956
0.2966
0.2968
0.3046
0.2938
0.2917
0.2952
0.2957
0.2979
+
N
Chloride
Formate
CH2
2-Methoxyacetate
Propionate
Acetate
AlMeIm+
CH3
Benzoate
N
CH3
Chloride
+
N
Dimethylphosphate
Formate
CH2
2
-Methoxyacetate
Propionate
0.864
0.882
+
AlMe Im
2

New Journal of Chemistry
Page 6 of 25
[
6]
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
Table 1, continue.
ET(WB);
V ;
M
Cation
Anion
SA
SB
SD
SP
n
f(n)
3
kcal/mol
cm /mol
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
Acetate
Formate
58.91
58.27
58.07
0.051
0.052
0.054
0.891
0.828
0.873
1.149
1.123
1.026
0.824
0.823
0.834
217.71
201.15
243.50
1.4961
1.4974
1.4952
0.2909
0.2907
0.2921
N
+
2-Methoxyacetate
Propionate
HN
5
8.33
0.048
0.905
1.007
0.896
239.24
1.5070
0.2996
DBUH+
+
H C
NH2
3
N
H C
N
CH3
3
Acetate
57.96
0.164
0.816
1.022
0.802
182.82
1.4862
0.2872
H C
3
TMGH+
a- The solvent parameters listed for the IL-DMSO mixtures (DMSO = 0.6) are: empirical polarity (E (WB)), Lewis acidity (SA), Lewis basicity
T
(
SB), dipolarity (SD), polarizability (SP), refractive index (n) and Lorentz-Lorenz refractive index function (n). The molar volume (V ) is
M
that of the pure IL.
-
.

Page 7 of 25
New Journal of Chemistry
View Article Online
DOI: 10.1039/D0NJ02258F
[
7]
1
2
3
4
5
6
7
8
9
2
.1.1- Synthesis of ionic liquids based on 1-methyl- and 1,2-dimethylimidazole.
We synthesized these ILs according to the general scheme-1; the example we show is that
of 1-allyl-3-methylimidazolium acetate (AlMeImAcO). The steps included: preparation of the IL
in chloride form (1-allyl-3-methylimidazolium chloride; AlMeImCl); transformation of ion-
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
1
exchange macroporous resin into acetate (resin-OH  resin-AcO); conversion of (AlMelmCl
AlMeImAcO).

R
H C
R
2
1. 1 h; stiring; room T.
CH
3
Cl
+
CH
3
N
N
–
N
N
A
B
H C
2
o
+
2. 2h; MO; 70 C; 50 W
Cl
H C
R
H C
R
2
2
–
Resin-AcO
+
CH
O
O
+
CH
3
3
–
N
N
N
N
Cl
CH3
R = -H; -CH3
Scheme-1: Synthesis of 1-methyl- and 1,2-dimethylimidazole-based ILs, via alkylation (A)
followed by ion exchange (B).
We synthesized AlMeImCl by microwave (CEM-Discover DU-8316) assisted reaction of 1-
methylimidazole with allyl chloride in acetonitrile (MeCN). A solution of allyl chloride (40.7 mL;
0
.5 mol) in 20 mL MeCN was added slowly (1 h) to a solution of 1-methylimidazole (42 mL; 0.5
mol) in 50 mL MeCN at room temperature. We heated the reaction mixture at 70 C for 2h (50
W irradiation power) and then removed MeCN. We purified the produced AlMeImCl by
vigorous agitation with hot ethyl acetate (3 x 40 mL each; 10 minutes), followed by phase
separation at room temperature (IL is the lower phase), and removal of residual ethyl acetate
under reduced pressure. Yield= 90%.
We converted the above-mentioned IL into the corresponding acetate using Amberlite IRN-
78, 1.2 mmol OH/mL resin. First, the resin was converted into the acetate form by agitation
with an aqueous solution of acetic acid (5 mol % excess acid) for 2 h. We filtered the resin, and

New Journal of Chemistry
Page 8 of 25
View Article Online
DOI: 10.1039/D0NJ02258F
[
8]
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
suspended it in water several times to remove the excess acid (vigorous agitation, 10 minutes
each), and then conditioned the resin in methanol. We dissolved 31.73g AlMeImCl (0.2 mol) in
500 mL methanol and passed the solution slowly through a column containing 250 mL of Res-
AcO, followed by 250 mL of methanol. We checked the completeness of ion exchange by
adding aliquots of the eluted solution to AgNO /HNO solution; no precipitate was formed. We
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
3
removed methanol, and then dried the obtained IL under reduced pressure, over P O until
4
10
constant mass; yield 95%. The other IL-carboxylates were obtained by a similar protocol, i.e.,
using Resin-PrO, etc. All ILs are slightly yellowish viscous liquids or soft pastes; they gave
1
satisfactory H NMR data; vide Table ESI-1.
2
.1.2- Synthesis of ionic liquids based on DBU and TMG.
We synthesized these ILs as given elsewhere22,27 by carefully neutralized a solution of the
super-base (0.2 mol/25 mL MeCN) with the appropriate carboxylic acid (0.2 mol/25 mL MeCN),
o
under nitrogen at 0 C, followed by agitation for 1h at room temperature. We removed MeCN
and dried the IL under reduced pressure, over P O until constant mass; yields 87 to 93%. The
4
10
DBU-based ILs are pale-yellow liquids; TMG-based IL is a white solid.
2
.1.3- Synthesis of 1-allyl-2,3-dimethylimidazolium dimethylphosphate (AlMe ImMe P)
2
2
2
4
We synthesized this IL according to scheme-2 below. The product is a pale-yellow liquid
that gave satisfactory 1H NMR spectrum, see Table ESI-1. We obtained 1-allyl-2-
methylimidazole in 55 % yield and AlMe ImMe P in 80% yield.
2
2

Page 9 of 25
New Journal of Chemistry
View Article Online
DOI: 10.1039/D0NJ02258F
[
9]
1
2
3
4
5
6
7
8
9
CH3
N
CH3
1. 6h; refluxed
–
+
H C
2
N
CH
N
N Na
+ NaCl
2
+
Cl 2. Stiring overnight; room T
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
H C O
3
O
P
CH
3
CH3
O
H C O
–
3
O
O
1. 1h, room T
2. MO, 2h, 50 W, 70 C
CH
P
3
H C
+
3
N
N
O
o
CH
2
O
CH
3
Scheme 2: Synthesis of 1-allyl-2,3-dimethylimidazole dimethylphosphate from 1-allyl-2-
methylimidazole followed by its alkylation by trimethylphosphate.
2
.2- Solvatochromic indicators (probes)
We show in Figure 1 the solvatochromic indicators employed in the present work, along
with other probes that we will discuss. We attach to each probe (or pairs of probes) the solvent
descriptor that it is used to calculate. These are referred to as (where S is the solvent): empirical
polarity (E (RB), E (WB)), E (MePMeBr ); solvent Lewis acidity, SA; solvent Lewis basicity, SB;
T
T
T
2
solvent dipolariy, SD; solvent polarizability, SP.

New Journal of Chemistry
Page 10 of 25
View Article Online
DOI: 10.1039/D0NJ02258F
[
10]
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 1: Molecular structures of the solvalchromic probes of interest in the present work.
WB: 2,6-dichloro-4-(2,4,6-triphenyl-1-pyridinio)phenolate; RB: 2,6-diphenyl-4-(2,4,6-
triphenylpyridinium-1-yl) phenolate; MePMeBr : 2,6-dibromo-4-[(E)-2-(1-methylpyridinium-
2
4-yl)ethenyl] phenolate; TBSB: o-tert-butylstilbazolium betaine; DTBSD: o,o’-di-tert-
II
butylstilbazolium betaine; FePhen: [Fe (1,10-phenanthroline) (CN) ]; NI: 5-nitroindoline;
2
2
MNI: 1-methyl-5-nitroindoline; DMANF: 2-(N,N-dimethylamino)-7-nitrofluorene; β-

Page 11 of 25
New Journal of Chemistry
View Article Online
DOI: 10.1039/D0NJ02258F
[
11]
1
2
3
4
5
6
7
8
9
carotene: (all trans) 1,10-(3,7,12,16-tetramethyl-1,3,5,7,9,11,13,15,17-octadecanonaene-
,18-diyl)bis[2,6,6-trimethylcyclohexene].
1
RB (Reichardt betaine), NI, anthracene, and -carotene were commercial (Aldrich), DMANF,
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
MNI, DTBSB, TBSB, FePhen and WB (Wolfbeis betaine) were available from previous studies.28-
3
3
2
.3- Solvatochromic responses of the dyes: Uv-Vis spectroscopy
2.3.1- Preparation of the binary IL-DMSO mixtures
o
We weighed each IL in a volumetric flask; dried it under reduced pressure at 60 C for
6
h, and weighted it again. The mass of dry DMSO required to obtain DMSO = 0.6 was then
introduced, and the mixture homogenized by (vortex) agitation.
2
.3.2- Preparation of probe solutions in the IL-DMSO binary mixtures; Uv-Vis spectra recording,
and refractive index measurements.
Stock solutions of the probes (0.036 mol/L) were prepared in acetone. We added 80 L
(most probes) or 150 L (FePhen; WB) of the probe solution to glass vials with threaded capes
(VWR 470151-622); evaporated the acetone; added 1 mL of the appropriate BM and dissolved
-3
the solid probe using a vortex mixer. The final probe concentrations were 2.88 x 10 , for all
-3
probes; 5.4 x 10 mol/L, for FePhen and WB.
We recorded each spectrum twice using Shimadzu UV-2550 UV-Vis spectrophotometer
o
under the following conditions: Temperature 40 ± 0.05 C (4000A digital thermometer, Yellow
Springs Instruments), scanning rate = 160 nm/min; 0.5 cm path length cuvettes with PTFE
stoppers. We calculated the values of max of the solvatochromic peaks (i.e., longest
wavelength) from the first derivative of the spectra (OriginLab 8 software). The uncertainty in
max was ≤ 0.5 nm, leading to ± 0.06 kcal/mol uncertainty in E (WB) (spectral range = 470 to 510
T
nm); ± 0.004 in SA (spectral range = 590 to 610 nm); ± 0.02 in SB (spectral range = 420 to 460
nm); ± 0.02 in SD (spectral range = 435 to 460 nm); ± 0.01 in SP (spectral range = 379 to 392
nm).

New Journal of Chemistry
Page 12 of 25
View Article Online
DOI: 10.1039/D0NJ02258F
[
12]
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
We used Rudolph Research J357 digital refractometer (operating at 488 nm) to measure

the refractive indices of IL-DMSO BMs (DMSO = 0.6) at 40 ± 0.05 C.
2
.4- Calculations
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
.4.1- Calculations of the solvatochromic parameters of the IL-DMSO binary mixtures
These were calculated as given elsewhere. We list in Table 2 each solvatochromic property
(
descriptor), the equation employed, and the corresponding reference. Here max corresponds
to the wavelength maximum of the solvatochromic peak of the probe;  is the corresponding
-1
frequency in cm .
Table 2: List of the solvatochromic probes along with the equations employed to calculate the
corresponding solvent descriptors.a
Solvent
Equation
Probe
Equation
Reference
descriptor
RB,
WB
ET(RB)
ET(WB)
3
4
1
2
28591.5 /max
35
MePMeBr2
TBSB
E (MePMeBr )
T
2
0.4 Δ/ 1299.8
Δ = TBSB – (1.4049 DTBSB – 6288.7)
(max – 16255)/3637
SA
36
DTBSB
3
4
FePhen
NI, MNI
DMANF
Anthracene
SA
SB
37
29
( – MNI) – 1570/ – 1735
NI
(solvent – DMANF)/1611
solvent = 28224.6 – 4887.45 SP
(24402 – β-carotene)/3672
(27660 – anthracene)/1612.7
5
SD
38
6
7
-carotene
SP
SP
39
38
Anthracene
a- Values of E (RB) and E (WB) are in kcal/mol.
T
T
2.4.2-Caculation of the molar volumes (V ) of the neat ionic liquids
M

Page 13 of 25
New Journal of Chemistry
View Article Online
DOI: 10.1039/D0NJ02258F
[
13]
1
2
3
4
5
6
7
8
9
The values of V were calculated using the Molinspiration platform.40 To ensure the
M
reliability of the volumes calculated by this software, we calculated V for 16 ILs (Table ESI-2),
M
based on their molar mass and density41,42 and correlated these (experimentally based) values
with those calculated by the above-mentioned software. We obtained the following equation;
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
showing the reliability of the theoretically calculated values of VM.
2
VM;Experimental = 5.94866 + 1.04052 VM;Calculated
N = 16; R : 0.99123
Equation 8
3
- Results and Discussion
3.1- Relevance of solvatochromic data to dissolution of carbohydrates
The term solvatochromism refers to the effects of the solvent on the spectra, absorption or
emission, of solutes (e.g., the probes shown in Figure 1) that are sensitive to a particular solute-
solvent interaction. Effects of solvents, or solvent mixtures, on the values of max of the
34
solvatochromic peak is then manipulated to get information on SA, SB, … etc, see Table 2. As
Shown by Eq. 9, the solvent empirical polarity E (probe) is the sum of contributions from H-
T
bonding terms (SA and SB), dipolar- (SD), and dispersive interactions (SP).43
E (probe) = E (probe) + aSA + bSB + dSD + pSP
Equation 9
T
T
0
where the descriptors (SA, SB, SD, SP) are those defined above; a, b, d, and p are the
corresponding regression coefficients. For conformity with other terms in Eq. 9, we use the
4
3
symbol SD instead of SdP (used by Catalán), to denote solvent dipolarity.
In using Eq. 9 we are interested in the sign and magnitude of the regression coefficients.
The former indicates whether a phenomenon is favored or not (positive and negative sign,
respectively) by a specific solute-solvent interaction mechanism, e.g., H-bonding. Comparison
of the magnitudes of the regression coefficients shows their relative importance to the
phenomenon in question, e.g., solvatochromism.
As argued in Introduction, the interaction mechanisms that contribute to E (probe) are
T
precisely those considered relevant for the dissolution of carbohydrate polymers. This is the
underlying reason for using solvatochromic parameters for assessing/predicting the efficiency
of, e.g., ILs as carbohydrate solvents. In other words, information on the sign/magnitude of the

New Journal of Chemistry
Page 14 of 25
View Article Online
DOI: 10.1039/D0NJ02258F
[
14]
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
regression coefficients of Eq. 9 can be fruitfully employed to predict whether an IL, or a family
of ILs, should be good solvents for cellulose and other carbohydrate polymers.19,15 An example
of this prediction is that ILs with effective basicity (= SB – SA) in the range 0.35 < SB – SA < 0.9
are expected to be good cellulose solvents.44,45
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
.2- Choice of the solvatochromic probes
As shown in the authoritative book on solvents and solvent effects in organic chemistry,34
46
and in a more recent review on the subject, there is a myriad of probes employed for the
calculation of solvent descriptors. The probes chosen are usually a matter of suitability for the
determination, vide infra; convenience and, sometimes, availability. As an example, we
successfully used commercially available -carotene to calculate SP instead of the polyene
employed by Catalán whose synthesis involves 15 steps.39,47
Before discussing the regression data, we comment briefly on the rationale for choosing
the probes. We employed WB because its lower pKa in water than RB (= 8.65 and 4.78, for RB
and WB, respectively),32,34 makes it less prone to protonation, e.g. by traces of acidic impurity, if
present, in the protic ILs. Note that E (WB) and E (RB) are linearly correlated (correlation
T
T
coefficient = 0.9905, N = 32 MSs).32
To calculate specific properties, for example SA, a pair of homomorphic probes (e.g., TBSB
and DTBSB) is usually employed. The first, TBSB, is sensitive to solvent Lewis acidity, in addition
to non-specific interactions. The second, DTBSB, is sensitive only to non-specific interactions.36
The value of SA is calculated by manipulating the spectral data of this pair of probes. We were
unable to use TBSB to calculate SA because its dissolution in the BM resulted in disappearance
of its solvatochromic peak. Schade et al. (2014) indicated the instability of the same probe in
3
7
other imidazole-based ILs, due to an (unspecified) side reaction. In agreement with their
explanation; we found that the solvatochromic peak of TBSB was not recovered by addition of
triethylamine (to deprotonate the probe). For calculation of SA, we employed the probe FePhen
that is stable in acidic media. Note that calculated values of SA using FePhen are linearly
3
7
correlated with SA calculated using TBSB and DTBSB. Solubility problem of -carotene in some
3
8
of these BMs precluded its use to calculate SP. As shown elsewhere, anthracene can be used

Page 15 of 25
New Journal of Chemistry
View Article Online
DOI: 10.1039/D0NJ02258F
[
15]
1
2
3
4
5
6
7
8
9
for this determination. Catalán and Hopf (2004) showed that Anthracene shows a linear
correlation with the Lorentz-Lorenz function of the refractive index (n). This function, on the
48
other hand correlates with SP, i.e., Anthracene correlates with SP. Calculation of SD requires use
of spectral data of DMANF and a polarizability probe, in our case anthracene.38
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
.3- Correlation of the solvatochromic descriptors of IL-DMSO mixtures
Based on the data of Table 1, we calculated the above-mentioned descriptors for IL-DMSO
o
mixtures at 40 C and fixed 
= 0.6. Table 3 shows the correlations calculated. Note that the
DMSO
solvent descriptors employed in these correlations are in their reduced form, as shown by Eq.
10 below, where Solv = solvatochromic. Consequently, the values of every parameter vary from
0 to 1, i.e., we can directly compare the magnitudes of the regression coefficients. In the
following discussion, we concentrate on part I of Table 3; we discuss part II of the same table
later.
Equation 10

New Journal of Chemistry
Page 16 of 25
[
16]
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
Table 3. Correlations of the empirical polarity E (probe) with the solvatochromic parameters SA, SB, SD, SP; and other solvent
properties: the molar volume of neat IL (V ); the Lorentz-Lorenz refractive index function (n) of IL-DMSO. Part (I) is for ionic liquids-
T
M
DMSO; Part (II) is for neat molecular solvents.
2
R ;
Number of
solvents
Entry
Correlated parameters
Correlation equation
Q2

0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
I- Correlations of the solvatochromic probes in ILs-DMSO,a
ET(WB) vs. SA, SB,
SD, SP
0.533
9.45
1
2
3
4
18
18
13
13
ET(WB) = +60.03(0.87) – 1.39(0.66) SA – 0.34(0.65) SB – 1.65 (0.81) SD – 1.79(0.85) SP
ET(WB) vs. SA, SB,
VM, (n)
0.567
8.52
E (WB) =+58.18(0.65) – 2.15(0.71) SA – 0.20(0.62) SB - 2.10(0.88) V – 0.70(0.94) (n)
T
M
ET(WB) vs. SA, SB,
SD, SP
0.889
1.133
0.939
0.593
ET(WB) =+54.28(0.70) + 3.74(0.59) SA + 0.86(0.35) SB + 0.52(0.49) SD + 0.19(0.46) SP
E (WB) =+54.61(0.28) + 2.77(0.37) SA + 0.61(0.24) SB - 1.06(0.34)) V + 0.41(0.33) (n)
ET(WB) vs. SA, SB,
VM, (n)
T
M
II- Correlations of the solvatochromic probes in neat molecular solvents,b
ET(WB) vs. SA, SB,
VM, (n)
0.820
116.7
0.885
77.7
5
6
7
23
23
23
E (WB) = +45.61(3.10) + 13.25(2.42) SA + 3.65(2.73) SB - 6.49(3.71) V + 0.42(2.87) (n)
T
M
E (RB) vs. SA, SB, V ,
T
M
E (RB) = +36.81(2.45) + 14.17(1.91) SA + 3.82(2.16) SB - 4.90(0.88) V + 0.70(2.26) (n)
T
M

(n)
E (MePMBr ) vs. SA,
0.817
47.24
T
2
E (MePMBr ) = +43.45(1.98) + 5.86(1.54) SA + 5.49(1.74) SB - 7.78(2.36) V + 2.29(1.82) (n)
T
2
M
SB, V , (n)
M

Page 17 of 25
New Journal of Chemistry
[
17]
1
2
3
4
5
6
7
8
9
2
2
a- We calculated the solvatochromic data for the binary mixtures from measurements at 40 C, at DMSO = 0.6. R and Q refer to
the correlation coefficient, and the sum of the squares of the residuals, respectively.
b- Regression analysis based on the solvatochromic data of Table ESI-3, at 25 C in. The values of V and (n) were calculated as
M
given for ILs, and ILs-DMSO, respectively.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8

New Journal of Chemistry
Page 18 of 25
View Article Online
DOI: 10.1039/D0NJ02258F
[
18]
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
We have two observations regarding entry 1 of Table 3 that shows the results of
2
2
correlation of E (WB) with the four solvatochromic probes for 18 BMs: the values of R and Q
T
are modest; except for the first term, E (WB) , the signs of the solvent descriptors are negative.
T
0
The first result may be due to inclusion of three classes of protic and aprotic ILs in the
correlation. These are based on substituted imidazole, DBU and TMG. As argued elsewhere for
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
4
9
an extensive series of neat ILs, good correlations are obtained for families with the same
cation (e.g., those based on imidazole) and different anions. The negative sign of all correlation
coefficient cannot, however, be readily explained. Consider the effect of solvent Lewis acidity
on the value of E (WB). For imidazole-based ionic liquids C2-H of the diazole ring is relatively
T
5
0
1
13
acidic, and forms H-bonds with the phenolate oxygen of WB, as shown by H and C NMR.
5
1,52
-
+
This H-bond increases the energy of the intramolecular charge transfer in WB (O  N ),
leading to a hypsochromic (i.e., a blue) shift of the corresponding peak, with a concomitant
increase in E (WB). Therefore, an increase in SA of the solvent increases the value of E (WB),
T
T
consequently the sign of the SA term should be positive; this is not the case (in entry 1 of Table
3).
Another factor that bears on the quality of the correlation is the uncertainty in the
calculated solvatochromic parameters. This is a consequence of the magnitude of  (= largest
max - smallest max) for a given probe within the 18 BMs studied (Table ESI-4). Values of  in
nm are: 32.3, 16.0, 34.0, 29.4, and 3.1 for WB, FePhen, NI, MNI, and anthracene, respectively.
The small  value for last probe and the fact that values of max for 10 BMs lie between 380
and 381 nm, prompt us to verify whether the use of another physicochemical property related
to SP would result in a better correlation.
3
9
53,54
Solvent effects on  of -carotene (used for calculation of SP) in organic solvents,
and BMs55,56 showed linear correlation with the solvent Lorenz-Lorentz refractive index function
2
2

(n) ( = (n -1)/(n +2)) , as shown in part A of Figure ESI-1. As mentioned earlier, there is a linear
4
8
correlation between Anthracene and (n). Consequently, we determined the refractive indices
of the BMs, and calculated (n). As the correlation of Anthracene and (n) for the BMs is also
linear (part B of Figure ESI-1), we decided to test this property instead of SP.

Page 19 of 25
New Journal of Chemistry
View Article Online
DOI: 10.1039/D0NJ02258F
[
19]
1
2
3
4
5
6
7
8
9
As shown by entry 5 of Table 2, we use SP to calculate SD; any uncertainty in the former
is carried over to the latter. Therefore, we sought another physiochemical property to use in
the correlation, instead of SD. For a homologous series of ILs (1-alkyl-3-methyimidazolium
bis(trifluoromethylsulfonyl)amide, alkyl = n-propyl, n-butyl, n-octyl, n-decyl), the values of
E (RB) decrease as a function of increasing the length of the attached alkyl group (E (RB) = 51.9;
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
T
T
5
1.5; 51.1; 51.0, respectively), i.e., as a function of increasing the molar volume of the solvent,
57
V . Similar phenomena were observed for cellulose swelling by MSs, and its dissolution in ILs,
M
two processes that are sensitive to the same biopolymer-solvent interactions as the
solvatochromic probes, vide supra. Thus, cellulose swelling by a series of aliphatic alcohols
(ethanol to n-octanol) and 2-alkoxyethanols (2-methoxyethanol to 2-(n-butoxy) ethanol
decreases regularly (exponential decay) as a function of increasing V . Inclusion of the latter
M
parameter in the correlation of cellulose swelling (wt%) with solvent descriptors lead to better
58
correlations. Likewise, cellulose dissolution (wt%) decreases as a function of increasing the
4
values of V for a series of homologous DBU carboxylates, and 1,3-disusbtituted imidazolium
M
2
1
acetates. Therefore, V is related to both E (RB) and cellulose dissolution efficiency. Note that
M
T
values of V are related to solvent SD, as follows: For a series of MSs (Table ESI-5), there is a
M
correlation between values of SD and the relative permittivity , see Part A of Figure ESI-2. For
the same MSs, and also for a series of ILs (Table ESI-6) there is an inverse relationship between

and V , (see Parts B and C of Figure ESI-2).59,60 Therefore, there is theoretical ground for
M
employing V and (n) in the correlations instead of SD and SP, respectively.
M
Entry 2 of Table 3 shows that this substitution resulted in a slightly better correlation for
the 18 BMs; the sign of the V descriptor is negative, as expected, based on the preceding
M
discussion (effect of V on E (probe)). As entry 3 of Table 3 shows, however, the quality of the
M
T
regression increased noticeably when we restricted the ILs to the 13 imidazole-based series,
and substantially when we used VM and (n) instead of SD and SP, respectively. More
importantly, however, the signs of all descriptors in entries 3 and 4 are the expected ones. That
is, the values of the empirical polarity of the BMs increase as a function of:
 Increasing the Lewis acidity, due to H-bonding of the solvent to the phenolate oxygen of
5
0
WB in agreement with the results of NMR data, and theoretical calculation of the

New Journal of Chemistry
Page 20 of 25
View Article Online
DOI: 10.1039/D0NJ02258F
[
20]
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
61
solvation of the RB probe. Catalán also showed a similar sensitivity to solvent Lewis
acidity;43
 Increasing the Lewis basicity, due to electrostatic attraction of the acetate ion to the
positive nitrogen of WB;
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0


Increasing SD, SP and (n).
Decreasing the molar volume of the solvent VM.
Values of the regression coefficients indicate that solvent Lewis acidity is the dominant
factor that affects its empirical polarity. The lower value of the regression coefficient of SB is
presumably due to the fact that the quaternary nitrogen of WB (also the structurally similar RB)
6
1
62
is sterically crowded, and its charge is diffused in the ring. We corroborate this conclusion
convincingly by the regression data of entries 5 to 7 of Table 3. In these, we carried out
correlations for 23 MSs, using 3 probes, WB and RB of similar steric crowding around the
probe´s positive nitrogen (by two phenyl groups), and MePMeBr where this crowding is
2
absent. The ratios of the regression coefficients (b/a, Eq. 9) are 0.28, 0.27, and 0.94 for WB, RB,
and MePMeBr , respectively. Based on this set of solvents, it is safe to conclude that the low
2
sensitivity towards SB in entries 1-4 of Table 3 is steric hindrance, i.e., solvent basicity is also
important to solvatochromism, hence to biopolymer dissolution, in agreement with published
1
7
data. Entries 3 and 4 of Table 3 shows the importance of solvent polarizability.
Taken together, these results confirm that solvent descriptors that affect E (probe) are the
T
same that are relevant to cellulose dissolution. Consequently, using solvachromic descriptors as
guidelines for assessing solvent efficiency (for carbohydrate dissolution) is justified. With this
relationship confirmed, we are planning to test the correlation between these parameters and
cellulose dissolution efficiency by the same binary solvent mixtures.
4
-Conclusions
The aim of this work is to contribute to the understanding of solvent efficiency in dissolving
cellulose. and carbohydrates in general. Because solvatochromism is sensitive to the same
solute-solvent interactions that control carbohydrate dissolution, in particular, H-bonding and
hydrophobic interactions, we correlated the empirical (or overall) solvent polarity, E (probe),
T

Page 21 of 25
New Journal of Chemistry
View Article Online
DOI: 10.1039/D0NJ02258F
[
21]
1
2
3
4
5
6
7
8
9
and the descriptors of specific solute-solvent interactions. In agreement with previous data on
pure ILs,37,49 we found satisfactory correlations between E (WB) and the descriptors of IL-DMSO
T
binary mixtures, especially if the ILs tested are limited to the same class of compounds, derived
from 1-methyl- and 1,2-dimethylimidazole (13 ILs). The quality of the correlation is enhanced by
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
keeping the H-bonding terms and substituting SD by V , and SP by (n). This substitution
M
eliminates the inherent uncertainties associated with calculation of SP, hence of SD. Whereas
SA is important in all correlations, the relative importance of SB is underestimated because the
positive nitrogen atoms in RB and WB are sterically crowded. We corroborated this conclusion
by correlations of three probes in 23 MSs, WB, RB (steric crowding) and MePMeBr (free of
2
crowding). In summary, our results indicate that efficient solvents for cellulose, and presumably
for chitin and starch, should be dipolar and efficient Lewis acids and Lewis bases. There is a
delicate balance between the hydrophobicity of the cation and its molar volume that controls
its proximity to the probe. Using solvatochomism for solvent efficiency screening for
carbohydrate dissolution is an excellent choice. It saves labor and cost because, unlike the
laborious biopolymer dissolution experiments, the solvatochromic measurements are relatively
simple, accurate and require much less material.
Acknowledgments
O. A. El Seoud thanks FAPESP (São Paulo State Research Foundation) for financial support
(grant 2014/22136-4) and CNPq (National Council for Research) for research fellowship (grant
306108/2019-4). T. A. Bioni thanks CAPES (Federal Agency for Graduate Studies) for PhD
fellowship (Grant 88882.328217/2019-01; code 001); M. T. Dignani thanks FAPESP for an
undergraduate research fellowship (grant 2019/02928-7). We thank G. El Seoud for drawing the
Table of Contents Entry.
5
-References
1
2
E. Borbély, Acta Polytech. Hung., 2008, 5 (3), 11-18.
C. L. Mccormick, P. A. Callais, B. H. Hutchinson, Macromol, 1985, 18 (12), 2394-2401.

New Journal of Chemistry
Page 22 of 25
View Article Online
DOI: 10.1039/D0NJ02258F
[
22]
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
3
T. Heinze, S. Köhler, In Cellulose Solvents: For Analysis, Shaping and Chemical Modification, ed.
T. Liebert, T. Heinze, E. J. Kevin, American Chemical Society, Washington, 1033, 2010,
Chapter 5, 103-118.
4
5
H. Hanabusa, Y. Takeoka, M. Rikukawa, M. Yoshizawa-Fujita, Aust. J. Chem., 2019, 72 (2), 55-
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
6
0.
D. C. Ferreira, M. L. Oliveira, T. A. Bioni, H. Nawaz, A. W. T. King, I. Kilpelainen, M. Hummel, H.
Sixta, O. A. El Seoud, Carbohyd. Polym., 2019, 212, 206-214.
6
7
8
Y. Huang, P. Xin, J. Li, Y. Shao, C. Huang, H. Pan, ACS Sustain. Chem. Eng., 2016, 4, 2286-2294.
A. Xu, J. Wang, H. Wang, Green Chem., 2010, 12 (2), 268-275.
J. Zhang, L. Xu, J. Yu, J. Wu, X. Zhang, J. He, J. Zhang, Sci. China Chem., 2016, 59 (11), 1421-
1
429.
9
O. Kuzmina, J. Bhardwaj, S. Vincent, N. Wanasekara, L. Kalossaka, J. Griffith, A. Potthast, S.
Rahatekar, S. Eichhorn, T. Welton, Green Chem., 2017, 19 (24), 5949-5957.
10
L. A. Ramos, E. Frollini, A. Koschella, T. Heinze, Cellulose, 2005, 12, 607–619.
1
1
G. T. Ciacco, D. L. Morgado, E. Frollini, S. Possidonio, O. A. El Seoud, J. Bra. Chem. Soc., 2010,
2
1, 71–77.
1
1
2
3
Y. Cao, J. Wu, T. Meng, J. Zhang, Carbohyd. Polym., 2007, 69, 665–672.
M. Abe, K. Sugimura, Y. Nishio, ChemistrySelect, 2016, 1, 2474–2478.
T. Heinze, O. A. El Seoud, A. Koschella, Cellulose Derivatives. Synthesis, Structure and
Properties. Springer International Publishing, Berlim, 2018.
14
1
1
1
5
6
7
M. Kostag, M. Gericke, T. Heinze, O. A. El Seoud, Cellulose, 2019, 26, 139-184.
F. M. Hämmerle, Lenzinger Ber., 2011, 89, 12-21.
O. A. El Seoud, M. Kostag, K. Jedvert, N. I. Malek, Macromol. Mater. Eng., 2020, DOI:
1
0.1002/mame.2019008.
18
B. Lindman, G. Karlström, L. STIGSSON, J. Mol. Liq., 2010, 156, 76–81.
O. A. El Seoud, M. Kostag, K. Jedvert, N. I. Malek, Polymers, 2019, 11, 1917-1921.
M. Clough, Green Chem., 2017, 19 (20), 4754-4768.
1
2
2
9
0
1
B. Lu, A. XU, J. WANG, Green Chem., 2014, 16 (3), 1326-1335.

Page 23 of 25
New Journal of Chemistry
View Article Online
DOI: 10.1039/D0NJ02258F
[
23]
1
2
3
4
5
6
7
8
9
22
A. King, J. Asikkala, I. Mutikainen, P. Järvi, I. Kilpeläinen, Angew. Chem., 2011, 50 (28), 6301-
305.
6
2
3
M. Kostag, M, T. Dignani, M. C. Lourenço, T. A. Bioni, O. A. El Seoud, Holzforschung, 2019, 73,
1103–1112.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
2
4
5
D. Zhao, H. Li, J. Zhang, L. Fu, M. Liu, J. Fu, P. Ren, Carbohyd. Polym., 2012, 87, 1490-1494.
A. Xu, Y. Zhang, Y. Zhao, J. Wang, Carbohyd. Polym., 2013, 92, 540-544.
W. L. F. Armagero, C. L. L. Chai, Purification of laboratory chemicals. Elsevier, New York, 2016.
V. Losetty, P. Matheswaran, C. Wilfred, J.Chem. Thermodyn., 2017, 105, 151-158.
G. Saroja, Z. Pingzhu, N. P. Ernsting, J. Liebscher, J. Org. Chem., 2004, 69, 987-990.
J. Catalán, C. Díaz, V. López, P. Pérez, J. De Paz, J. Rodríguez, Liebigs Annales, 1996, 11, 1785-
26
27
2
2
8
9
1
794.
3
0
J. Catalán, P. Pérez, J. Elguero, W. Meutermans, Chem. Ber., 1993, 126, 2445-2448.
A. A. Schilt, J. Am. Chem. Soc., 1960, 82 (12), 3000-3005.
31
32
C. T. Martins, M. S. Lima, O. A. El Seoud, J. Org. Chem., 2006, 71, 9068-9079.
O. A. El Seoud, Pure Appl. Chem., 2009, 81, 697-707.
3
3
3
3
3
4
5
6
C. Reichardt, Solvents and Solvent Effects in Organic Chemistry, Wiley-VCH, Weinheim, 2002.
M. A. Kessler, O. S. Wolfbeis, Chem. Phys. Lipids, 1989, 50 (1), 51-56.
J. Catalán, C. A. Díaz, Liebigs Ann., 1997, 1997 (9), 1941-1949.
37
38
A. Schade, N. Behme, S. Spange, Chem.: Eur. J., 2014, 20 (8), 2232-2243.
J. C. Del Valle, B. F. García, J. Catalán, J. Phys. Chem. B, 2015, 119 (13), 4683-4692.
C. Loffredo, P. Pires, M. Imran, O. A. El Seoud, Dyes Pigments, 2013, 96 (1), 16-24.
Molispiration Cheminformatics, http://www.molispiration.com, (accessed October 2019).
S. Zhang, N. Sun, X. He, X. Lu, X. Zhang, J. Phys. Chem. Ref. Data, 2006, 35 (4), 1475-1571.
G. R. Vakili-Nezhaad, A. M. Alaisaee, M. A. Aljabri, S. S. Albarwani, Z. K. Aljahwari, Mater.
Phys. Mech., 2017, 32, 8-13.
3
4
4
4
9
0
1
2
4
4
3
4
J. Catalán, J. Phys. Chem. B, 2009, 113 (17), 5951-5960.
L. K. J.Hauru, M. Hummel, A. W. T. King, I. Kilpeläinen, H. Sixta, Biomacromol., 2012, 13 (9),
2
896-2905.

New Journal of Chemistry
Page 24 of 25
View Article Online
DOI: 10.1039/D0NJ02258F
[
24]
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
45
A. Parviainen, A. King, I. Mutikainen, M. Hummel, C. Selg, L. Hauru, H. Sixta, I. Kilpeläinen,
ChemSusChem, 2013, 6 (11), 2161-2169.
4
4
6
7
V. G. Machado, R. I. Stock, C. Reichardt, Chem. Rev., 2014, 114 (20), 10429-10475.
D. Klein, P. Kiliçkiran, C. Mlynek, H. Hopf, I. Dix, P. G. Jones, Chem. Eur. J., 2010, 16, 10507-
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
0522.
4
8
J. Catalán, H. Hopf, Eur. J. Org. Chem., 2004, 22, 4694-4702.
49
50
S. Spange, R. Lungwitz, A. Schade, J. Mol. Liq., 2014, 192, 137-143.
S. Sowmiah, V. Srinivasadesikan, M. C. Tseng, Y. H. Che, Molecules, 2009, 14 (9), 3780-3813.
J. G. Dawber, R. A. Williams, J. Chem. So. Faraday Trans., 1996, 82 (1), 3097-3112.
M. Afri, H. E. Gottlieb, A. A. Frimer, Can. J. Chem., 2014, 92 (2), 128-134.
A. B. Myers, R. R. Birge, J. Chem. Phys., 1980, 73, 5314-5321.
5
5
5
5
1
2
3
4
T. Abe, J-L. M. Abboud, F. Belio, E. Bosch, J. I. Garcia, J. A. Mayoral, R. Notario, J. Ortega, M.
Posés, J. Phys. Org. Chem., 1998, 11, 193-200.
55
P. M. Mancini, A. C. Pérez, L. R. Vottero, J. Solution Chem., 2001, 30, 695-707.
P. M. Mancini, C. G. Adam, G. G. Fortunato, L. R. Vottero, Arkivoc, 2007, 16, 266-80.
C. F. Poole, J. Chromatogr. A, 2004, 1037, 49-82.
5
5
5
6
7
8
O. A. El Seoud, L. C. Fidale, N. Ruiz, M. L. O. D’Almeida, E. Frollini, Cellulose, 2008, 15, 371-
3
92.
69
60
J. P. Liu, W. V. Wilding, N. F. Giles, R. L. A. Rowley, J. Chem. Eng. Data, 2010, 55, 41-45.
T. Singh, A. Kumar, J. Phys. Chem. B, 2008, 112 (41), 12968-12972.
6
1
C. Chiappe, C. Pomelli, : A semiempirical study. Theorical Chemistry Acconts, v. 131, n. 3, p. 1-
7
, 2012.
6
2
H. Weingärtner, Z. Phys. Chem., 2006, 220, 1395-1405.

Page 25 of 25
New Journal of Chemistry
View Article Online
DOI: 10.1039/D0NJ02258F
1
2
3
4
5
6
7
8
9
Table of contents for
Understanding the efficiency of ionic liquids-DMSO as solvents for carbohydrates: Use of
solvatochromic- and related physicochemical properties
Thaís A. Bioni, Mayara L. de Oliveira, Marcella T. Dignani, Omar A. El Seoud,*
Institute of Chemistry, the University of São Paulo, 748 Prof. Lineu Prestes Av.,
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
0
5508-000 São Paulo, SP, Brazil. Email (elseoud.usp@gmail.com)
Quantification of interactions of solvatochromic probes with ionic liquids/DMSO serves as
expedient approach for predicting solvent efficiency in dissolving carbohydrates