Successful application of an ionic liquid carrying the fluoride counter-ion in biomacromolecular chemistry: Microwave-Assisted acylation of cellulose in the presence of 1-allyl-3-methylimidazolium fluoride/DMSO mixtures

Article abstract of DOI:10.1002/mabi.201200296

The use of ionic liquids with fluoride anion (IL-F) is challenging because of side reactions. Neat 1-allyl-3-methylimidazolium fluoride (AlMeImF) is used as a solvent in microwave-assisted acylation of cellulose. The results are disappointing due to side reactions in the IL proper, and F^--mediated hydrolysis of the produced ester. A dramatic improvement is observed, when AlMeImF/DMSO mixture is employed. The results are comparable to those obtained when pure 1-allyl-3-methylimidazolium chloride is employed. FTIR spectroscopy shows that dissolving a carboxylic acid anhydride in IL-F leads to the formation of acyl fluoride. Thus ILs are far from being "spectator" solvents. The new approach (use of IL-F/DMSO) is attractive because of its efficiency, low cost, and applicability to the derivatization of any polymer.

Full text of DOI:10.1002/mabi.201200296

Full Paper
Successful Application of an Ionic Liquid
Carrying the Fluoride Counter-ion in
Biomacromolecular Chemistry:
Microwave-Assisted Acylation of Cellulose in
the Presence of 1-Allyl-3-methylimidazolium
Fluoride/DMSO Mixturesa
Romeu Casarano, Omar A. El Seoud*
The use of ionic liquids with fluoride anion (IL-F) is challenging because of side reactions.
Neat 1-allyl-3-methylimidazolium fluoride (AlMeImF) is used as a solvent in microwave-
assisted acylation of cellulose. The results are disappointing due to side reactions in the IL
proper, and F–-mediated hydrolysis of the pro-
duced ester. A dramatic improvement is observed,
when AlMeImF/DMSO mixture is employed. The
results are comparable to those obtained when
pure 1-allyl-3-methylimidazolium chloride is
employed. FTIR spectroscopy shows that dissol-
ving a carboxylic acid anhydride in IL-F leads
to the formation of acyl fluoride. Thus ILs are
far from being ‘‘spectator’’ solvents. The new
approach (use of IL-F/DMSO) is attractive because
of its efficiency, low cost, and applicability to the
derivatization of any polymer.
1. Introduction
that are employed in diverse applications (fibers, films,
membranes, etc.). Cellulose-containing biomass is being
intensely investigated as raw material forthe production of
bioethanol,[1,2] and other important chemicals.[3]
Cellulose is the most abundant biopolymer; it is the source
of many important derivatives, including esters and ethers
The strong inter- and intra-molecular hydrogen-bonding
withinthebiopolymerisresponsibleforitssemi-crystalline
structure.[4–6] Cellulose dissolution, needed for derivatiza-
tion under homogeneous conditions, requires disruption of
this hydrogen-bonding network. Combinations of strong
electrolytes and dipolar aprotic solvents, e.g., LiCl/N,N-
dimethylacetamide and quaternary ammonium fluoride
hydrates (R4NF ꢀ xH2O) in DMSO have been successfully
R. Casarano, O. A. El Seoud
Institute of Chemistry, University of Sa˜o Paulo, P. O. B. 26077,
05513-970 Sa˜o Paulo, S. P., Brazil
E-mail: elseoud@iq.usp.br
a Supporting Information for this article is available from the Wiley
Online Library or from the author.
Macromol. Biosci. 2013, 13, 191–202
ß 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
wileyonlinelibrary.com
DOI: 10.1002/mabi.201200296
191

R. Casarano, O. A. El Seoud
www.mbs-journal.de
employed as media for cellulose dissolution and subse-
quent derivatization.[7–12]
1.3. Fluoride-Ion-Mediated Side Reactions
Another possible side reaction, common to esterification
inR4NF ꢀ xH2O/dipolar aproticsolventsisFꢁ-mediatedester
hydrolysis, as observed for benzoates-,[39] and cellulose
esters.[40]
Recently, ionic liquids (ILs), in particular those based
on imidazole have attracted much attention as solvents
for chemical reactions, including derivatization of
several biopolymers under homogeneous reaction condi-
tions.[6,13–15] There are several reasons behind this intense
interest: ILs are composed of ions, so that no additional
electrolyte, e.g., LiCl, is required for biopolymer dissolution;
theyarepolar, chemicallyandthermallystable, andpossess
extremely low vapor pressure, a property that enhances
process safety; their properties can be ‘‘tailored’’ to the
researcher’s needs by a judicious choice of the constituent
ions.[16–20]
In the present work, we have synthesized the IL 1-allyl-3-
methylimidazolium fluoride, AlMeImF, and used it as
solvent for the (homogeneous) acylation of cellulose by
carboxylic acid anhydrides. The results were disappointing
in terms of the ester degree of substitution (DS). We have
attributed this result to side reactions, and raised the
following questions: (1) What are the side reactions in
case of this IL-F? (2) Is it possible to eliminate/attenuate
them, so that the high electronegativity of the fluoride ion
can be advantageously exploited for cellulose dissolution/
derivatization? (3) What is the efficiency of AlMeImF as
compared withthatof the(extensively employed) 1-allyl-3-
methylimidazolium chloride, AlMeImCl?
An important mechanism for cellulose dissolution
involves hydrogen-bonding between the hydroxyl
groups of the anhydroglucose units (AGUs), and the
counter-ion of the IL.[6,21,22] Consequently, cellulose dis-
solution is largely dependent on the basicity of the anion;
chloride, acetate, formate, and dialkylphosphate have been
successfully employed.[6,13] The structure of the cation also
plays a role, e.g., imidazole-based ILs with side-chains
carrying heteroatoms (oxygen) dissolve cellulose better
than those with saturated side-chains, although
exceptions are known.[6,13,17,23–27] In this regard, the
fluoride counter-ion, being small and of higher charge-
density is expected to be the best member among the
halogen series. A literature survey shows only few reports
on the use of ILs with fluoride counter-ion, LI-Fs, for the
dissolution of cellulose,[18,28,29] and, to our knowledge,
none on the derivatization of the dissolved biopolymer.
Several reasons can be advanced in order to explain this
obvious gap.
Our preliminary results have shown that neat AlMeImF
is inferior to AlMeImCl. Surprisingly, however, addition of
DMSO to the former not only solved the problem of low
efficiency, but also led to the use of much less IL. In fact, we
used AlMeImF as an ‘‘electrolyte’’ in DMSO, akin to others
employed, e.g., LiCl, or R4NF ꢀ xH2O. Thus the extra labor in
the synthesis of this IL-F (ion-exchange) is more than
compensated for by the higher efficiency of the resulting
solvent system. Acylation of microcrystalline-, or eucalyp-
tus cellulose in AlMeImF/DMSO yielded cellulose car-
boxylic esters with DS comparable to that obtained with
neat AlMeImCl. In contrast to AlMeImF, dilution of the
corresponding chloride with DMSO yielded cellulose
acetates with lower DS.
2. Experimental Section
1.1. Unavailability of the Starting Materials
Notes: (i) As shown below, we have employed two samples of IL-F:
one obtained without addition of DMSO; the other obtained by
adding this molecular solvent to methanolic AlMeImF solution,
before evaporation of the alcohol. These will be referred to as
AlMeImF; AlMeImF/DMSO, respectively; (ii) the composition of
AlMeImF/DMSO solution is given on the mole fraction scale; e.g.,
AlMeImF0.22/DMSO0.78 referstoasolutionwherethemolefractions
are 0.22 and 0.78, for AlMeImF and DMSO, respectively.
Many organic fluorides, the starting materials for the
syntheses of LI-Fs, are not commercially available. There-
fore, the synthetic schemes for obtaining these (second
generation) IL-Fs involve the synthesis of another IL-halide,
followed by either ion-exchange or the use of (expensive)
AgF.
1.2. Side Reactions in the IL-F Proper
2.1. Materials
In principle, there exists the possibility of several
F–-mediated side-reactions, including abstraction of
the relatively acidic C2-H, or other hydrogens of the
imidazolium moiety,[30–34] elimination reactions involving
the aliphatic side chain via Hoffman degradation, or an
ylide mechanism, akin to the E1cB counterpart, except
that the species produced by proton abstraction is a
zwitterion.[35–38]
The reagents were purchased from Acros, Aldrich, or Merck
and were purified, where required, as described elsewhere.[41]
Microcrystalline cellulose (MCC); Avicel PH 101 was from FMC
(Philadelphia; viscosimetric degree of polymerization, DPv ¼ 175).
Eucalyptus sheets (Lwarcel Cellulose and Paper Co., S˜ao Paulo) were
cut into stripes, grounded, sieved (100–200 mesh) and then
mercerized by treatment with NaOH under reducing conditions,
as given elsewhere; DPv ¼ 965.[27]
Macromol. Biosci. 2013, 13, 191–202
ß 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
192
www.MaterialsViews.com

Successful Application of an Ionic Liquid Carrying the Fluoride Counter-ion . . .
www.mbs-journal.de
DU-8316, CEM) under the following conditions: MW-power, 50 W;
reaction temperature, 70 8C; time, 75 min. Ethyl acetate (100 mL)
was then added; the mixture stirred for 10 min; cooled in dry ice/
ethanol bath (IL-Cl solidification), the supernatant discarded; this
procedure was repeated at least three more times. The product was
dissolved in methanol (80 mL) and stirred overnight with activated
carbon (5 g). After filtration, the methanol was removed by heating
the IL for ca. 4 h, 2 mmHg, 60 8C. The product is light yellowish-
colored liquid that solidifies very slowly on standing (see part A of
Figure S1; Figure S1 of Supporting Information); yield ¼ 90%. The
1H NMR spectrum of AlMeImCl (see Figure S2 of Supporting
Information) agrees with literature data.[45–47]
2.2. Equipment
The value of DPv was determined by using shear-dilution Cannon-
Fenske viscosimeter (Schott), inserted in Schott AVS 360 automatic
viscosity determination equipment. 1H and 19F NMR spectra were
recordedwith VarianInnova-300spectrometer(300 MHz for 1H);IR
spectra were recorded with Bruker Vector-22 Fourier transform
infrared (FTIR) spectrophotometer.
2.3. Material Characterization
The DS of cellulose carboxylic esters was determined either by the
solvatochromic indicator method;[42] or by titration;[43] the results
of both methods agree within ꢂ 0.05. DPv was determined (25 8C)
from the intrinsic viscosity of cellulose solution in CUEN/water
(1:1 v/v) according to the recommended procedure.[44]
2.5.2. Clꢁ/OHꢁ Ion Exchange and Neutralization of
AlMeImOH with HF
The counter-ion of the IL was converted into the corresponding
hydroxide by ion exchange on a macro-porous resin (Purolite
SGA55OOH, 1.10 equiv. OHꢁ/L), as follows: 1 L of a methanolic
solutionofAlMeImCl(0.25 mol)waspassedthroughaglasscolumn
containing the resin (400 mL). The completeness of the (Clꢁ/OHꢁ)
ion-exchange was checked by treating a sample of the eluent with
acidified AgNO3 solution. The pH of the hydroxide solution was
brought to ca. 7 (expanded scale pH-paper) by adding methanolic
HF solution. The methanol was removed by evaporation; the
pressurewasgraduallyreducedfrom200to10 mmHgat50 8C, then
to 2 mmHg, at 60 8C for ca. 4 h. A dark amber-colored liquid was
obtained, see Figure S1B of Supporting Information. Figure S3 of
Supporting Information shows the 1H NMR spectrum of the
product, along with peak attribution of AlMeImF.
2.4. Procedures for Spectroscopic Analyses
For 1H NMR measurements, the sample solution in DMSO-d6 was
prepared ca. 30 min before the analysis; TMS was employed as
internalreference;19FNMRchemicalshiftswerecalculatedrelative
to an external reference, 1 wt.-% trifluoroacetic acid solution in
DMSO-d6,containedinaWilmad-Labglass516-I-3coaxial-tube.The
IR spectra were recorded for the samples diluted with DMSO, by
using 0.025 mm CaF2 cell; 64 scans were added at 0.5 cmꢁ1 digital
resolution.
2.5. Synthesis of 1-Allyl-3-methylimidazolium
Fluoride in the Absence of DMSO
Because of the dark color of the IL-F produced (Figure S1-B of
Supporting Information), methanol evaporation has also been
carried out either at room temperature at 2 mmHg, or in the
presence of propyl 3,4,5-trihydroxybenzoate (propyl gallate; 3 mol-
% with respect to IL-F), at 60 8C for ca. 4 h, at 2 mmHg.
The synthesis was carried out as shown in Scheme 1.
2.5.1. Synthesis of AlMeImCl
A mixture of 1-methylimidazol (20 mL, 0.25 mol) and allyl chloride
(22.8 mL, 0.275 mol) was heated in an MW apparatus (Discover
2.6. Synthesis of 1-Allyl-3-methylimidazolium
Fluoride with Addition of DMSO
AlMeImF was obtained as given above (see Section 2.5); DMSO was
added before the evaporation of methanol from the IL-F solution,
i.e., after its neutralization with HF.[48] During this step, some
DMSOwasremovedalongwithmethanol(ꢃ10%,bycomparingthe
initial and final masses). The exact concentration of AlMeImF in
DMSO was calculated by 19F NMR spectroscopy, by using 3-nitro-
a,a,a-trifluorotoluene as an internal reference. For comparison,
AlMeImCl/DMSOwasalsoprepared, butDMSOwasaddedafterthe
evaporation of methanol from the IL-Cl solution.
2.7. Determination of the Stability of AlMeImCl,
AlMeImF, and AlMeImF/DMSO
1
2.7.1. By Employing H NMR
A
methanolic solution containing equimolar quantities of
Scheme 1. The three steps of the synthesis of AlMeImF: the
reaction of 1-methylimidazole with allyl chloride; (Clꢁ ! OHꢁ)
ion-exchange; the neutralization of the hydroxide counterpart IL
with HF.
AlMeImCl and benzaldehyde (0.11 mol) was prepared; methanol
was then evaporated as described in the synthesis of AlMeImF
(pressure reduced from 200 to 10 mmHg at 50 8C, then at 2 mmHg,
Macromol. Biosci. 2013, 13, 191–202
ß 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
193
www.MaterialsViews.com

R. Casarano, O. A. El Seoud
www.mbs-journal.de
60 8C for ca. 4 h); the resulting sample was analyzed by 1H NMR
(DMSO-d6). The same procedure was employed for AlMeImF, and
AlMeImF/DMSO-d6; deuterated solvent was employed instead of
(CH3)2SO.
The solution was kept under the above-mentioned conditions
during the required length of time, vide infra. The resultant was
addedtoethanol(400 mL),stirredfor30–60 min,atca.60 8C,andthe
precipitated cellulose ester centrifuged at 3 500 g. This washing
procedure was repeated three more times; the product was dried
under reducedpressure at 50–60 8C, for 48 h, and its DSdetermined.
2.7.2. By Employing FTIR
Four solutions in DMSO, containing the same concentration,
0.43 mol ꢀ Lꢁ1, were prepared: (A) benzaldehyde, (B) AlMeImF, (C)
AlMeImF/DMSO, and (D) AlMeImCl. Equal volumes of solutions A
plus B; A plus C; A plus D were added to glass vials, equipped with
screw-caps. The latter were tightly closed, the components mixed,
and kept at room temperature. The IR spectra of each solution were
recorded right after mixing, and after 5, 24, 48, and 72 h. The area of
the benzaldehyde yC¼O peak at 1 698 cmꢁ1 was calculated from the
1 720 to 1 680 cmꢁ1 section of the spectrum, by employing a
commercial software (Grams/32; version 5, Galactic Industries,
Salem).
3. Results and Discussion
The Results and Discussion part is organized as follows:
(i) The results of cellulose acylation in neat AlMeImF are
shown; (ii) Possible side reactions that occur in this IL are
investigated; (iii) A solution is given for the problem of
side reactions; (iv) Results of acylation of celluloses in
AlMeImF/DMSO are listed and compared with those for
neat AlMeImCl and AlMeImCl/DMSO; (v) Nature of the
acylating agent(s) is discussed; (vi) A tentative explanation
for the role of DMSO is given.
2.8. Effect of Heating Cellulose Triethanoate with
AlMeImF or AlMeImF/DMSO on the Degree of
Substitution of the Ester
Solutions of commercial cellulose triethanoate (CA 398-6; DS ¼ 2.7;
Eastman) were prepared by mixing the ester (0.55 g) with 11 g of
each of the following in one-neck round-bottom flasks: AlMeImF;
AlMeImF/DMSO; or AlMeImCl. Using MW-heating (30 W), each
solution was heated at 80 8C, for 3 h, under mechanical stirring
(model RW 20; IKA Labortechnik; 570 rpm). The resulting solution
was added to ethanol (400 mL); the suspension was stirred for 30–
60 min; theprecipitatedsolidwas centrifuged at3 500 g(IEC Centra
244 MP4R). This procedure was repeated three more times; the
product was dried under reduced pressure at 50–60 8C, for 48 h, and
its DS determined.
3.1. Cellulose Acylation in Neat AlMeImF
After dissolution of MCC in neat AlmeImF, or AlMeImCl
(the latter for comparison), ethanoic anhydride was added
and the MW-assisted reaction was carried out at 80 8C
and 30 W, for 1–6 h. Table 1 shows the reaction conditions
and values of the DS obtained. Relative to AlMeImCl, the
acetylation reaction in IL-F is inefficient in terms of the DS
obtained. As will be discussed below, attempts to solve this
problem(entries6and7ofTable1)didnotchangemuchthe
DS obtained.
2.9. Investigation of IL-F/Acid Anhydride Mixtures
by FTIR
3.2. Investigation of the Possible Side Reactions that
Occur During the Synthesis of AlMeImF
In glass vials provided with screw-caps, solutions of the following,
each 0.43 M in DMSO, were prepared: (A) AlMeImF/DMSO; (B)
ethanoic anhydride; (C) hexanoic anhydride. Mixtures were then
prepared by adding equal volumes of (A) plus (B), and (A) plus (C).
For each mixture, the IR spectra were recorded right after mixing,
5 h, and 24 h.
We have raised two possible reaction pathways that can
contribute to the disappointing results shown in Table 1,
namely: (1) fluoride-ion mediated de-acylation, akin to that
observed for benzoate esters in (C4H9)4NF ꢀ 3H2O (TBAF)/
DMSO,[39] and cellulose triethanoate in tetraallylammo-
nium fluoride (TAAF) ꢀ H2O/DMSO;[40] (2) side reactions
occurring in the IL-F proper.
2.10. Dissolution and Acylation of Cellulose in
AlMeImF and AlMeImF/DMSO
With regard to (1), the following experiment has been
carried out: A commercial sample of cellulose triethanoate
(DS ¼ 2.7)wasdissolvedinneatAlMeImForAlMeImCl;each
solution was stirred at 80 8C (MW; 30 W); the DS of the ester
was determined after 3 h. The results obtained were as
follows: in AlMeImCl, DS ¼ 2.6; in AlMeImF, DS ¼ 1.1.
Therefore, Fꢁ-mediated deacylation occurs in neat IL-F.
The mechanistic possibilities for this reaction include:
fluoride ion-mediated ester hydrolysis, where (Fꢁ) is acting
either as a general base (a), or nucleophile (b); Fꢁ-mediated
deacylation duetosidereactionthatgeneratean ylide(c), or
Cellulose (0.50 g) was added to 10 g AlMeImF, AlMeImF/DMSO,
AlMeImCl, or AlMeImCl/DMSO. Each mixture was mechanically
stirred(570 rpm)at80 8C, for20–40 min,byapplyingMW-radiation
(30 W). Cellulose dissolution was followed visually by employing a
lamp placed behind the reaction flask. MCC and eucalyptus
cellulose dissolve readily in AlMeImCl or AlMeImCl/DMSO;
resulting in clear solutions. On the other hand, solutions of MCC
in AlMeImF or AlMeImF/DMSO were translucent; addition of the
carboxylic acid anhydride resulted in the formation of clear
solutions, fast for MCC and slow for fibrous cellulose.
Macromol. Biosci. 2013, 13, 191–202
ß 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
194
www.MaterialsViews.com

Successful Application of an Ionic Liquid Carrying the Fluoride Counter-ion . . .
www.mbs-journal.de
Table 1. Results of the acetylation of MCC in AlMeImF, or AlMeImCl, for comparison.a)
Entry
Solvent/reaction
medium
Ac2O/AGUb)
Reaction time
[h]
DS of the
product
1
2
3
4
5
6
7
8
9
AlMeImF
AlMeImF
AlMeImF
AlMeImF
AlMeImF
AlMeImFc)
AlMeImFd)
AlMeImCl
AlMeImCl
3
6
6
6
6
6
6
3
6
1
1
2
3
6
3
3
3
3
0.2
0.3
0.3
0.4
0.2
0.4
0.4
0.95
2.95
a)Microwave-assisted acetylation (30 W, 80 8C); b)molar ratio between ethanoic anhydride (AC2O) and AGU; c)the methanolic solution was
evaporated at room temperature, under reduced pressure; d)propyl gallate (3 mol-%/mol AlMeImF) was added prior to evaporation of
methanol.
an anion(d). Recently, wehave shown thatcellulose acetate
hydrolysis occurs in TAAF/DMSO, where no Fꢁ-mediated
elimination occurs.[40] Therefore, (a) or (b) may occur
concomitant with (c) or (d).
Scheme 3 below. Recently, it has been postulated that de-
aylation of cellulose ester by TBAF/DMSO occurs via Fꢁ-
mediated elimination of H-CH2CO-OCell, followed by
deacetylation, due to the formation of ketene (CH2¼C¼O),
alternative (d).[49] Although further discussion of these
possibilities is outside the scope of the present work, all are
based on the high basicity of Fꢁ in DMSO.
Scheme 2 addresses the mechanistic possibilities (a) and
(b). Part (A) of this scheme shows that the counter-ion of the
IL is promoting ester hydrolysis by acting as a general base
for the attack of water. The source of water may be the
biopolymer itself (not dried), or the (hygroscopic) IL-F. Part
(B) of the same scheme shows the intermediate formed (via
attack of Fꢁ on the ester C¼O group) if the fluoride ion were
acting as a nucleophile. This alternative is most certainly
not operative because Fꢁ is a better leaving group from the
tetrahedral intermediate than (Cell-Oꢁ).
Probing the nature and extent of side reactions occurring
in the IL proper is important per se, and because this bears
on the potential use of IL-Fs as solvents for chemical
reactions, including those of cellulose and other biopoly-
mers. AlMeImF was synthesized from AlMeImCl via
(Clꢁ ! OHꢁ) anion-exchange, followed by neutralization
with HF, and methanol removal by two-step evaporation;
vide Experimental. As shown in Figure S1 of Supporting
Information both ILs differ in color; AlMeImCl is light-,
whereas AlMeImF is dark-amber. A more fundamental
difference, however, is the evolution of gas bubbles during
Alternative(c)involvestheattack onthe esterC¼Ogroup
by an ylide, formed via Fꢁ-mediated abstraction of C2-H of
the imidazolium ring, followed by elimination of (Cell-Oꢁ).
The mechanism of formation of this ylide is shown in
Scheme 2. Part (A) shows the fluoride ion-mediated cellulose ethanoate hydrolysis through a general base catalyzed attack of water on the
ester acyl group. Part (B) depicts the structure of the tetrahedral intermediate formed by nucleophilic attack of (Fꢁ) on the ester, if the
reaction involves nucleophilic catalysis.
Macromol. Biosci. 2013, 13, 191–202
195
www.MaterialsViews.com
ß 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

R. Casarano, O. A. El Seoud
www.mbs-journal.de
Scheme 3. Suggested mechanism for the reactions occurring in the IL-F/aldehyde mixture. The initial step is F–-mediated abstraction of
C2-H of the imidazolium ring, followed by nucleophilic addition of the ylide formed to the carbonyl group of benzaldehyde.
the evaporation of methanol under reduced pressure (at
2 mmHg). This is observed only for AlMeImF and may be
taken to indicate side reactions, an assumption corrobo-
rated by comparing the 1H NMR spectra of both ILs;
AlMeImF has several peaks not present in the spectrum of
the corresponding chloride (Figure S2 and S3 of Supporting
Information).
particular, there is no peak at ca. 6.3 ppm, attributed to
the AlMeImCl-derived ylide-aldehyde addition product.
Figure 2 shows the same result for the IL-F. The salient
feature is the presence of several additional peaks (as
compared with Figure 1), in particular the singlet at ca.
6.3 ppm. This confirms the formation, and subsequent
addition of the above-mentioned ylide to benzaldehyde.
The decrease in the intensity of the latter peak after
complete evaporation of methanol (compare the whole
spectrum and the inset in Figure 2) is a consequence of
the reversibility of the two reactions (H-abstraction
and addition to C6H5CH¼O), as shown in Scheme 3, and
discussed elsewhere.[30]
Two strategies have been attempted in order to
eliminate/attenuate the above-mentioned problem: use
of lower evaporation temperature; addition of an efficient
antioxidant. Although methanol evaporation at room
temperature has resulted in lighter-color product (orange),
the corresponding 1H NMR spectrum still showed
additional peaks (spectrum not shown). Because propyl
gallate is an efficient inhibitor for both homolytic and
heterolytic side reactions,[50] we tested the effect of
its addition before methanol evaporation. Again, the
IL-F produced was dark-colored (probably due to the
oxidation of propyl gallate);[50] evolution of gas has been
observed during the evaporation of methanol; 1H NMR
spectrum showed several additional peaks (spectrum not
shown). In summary, both procedures did not solve the
problem.
FTIR is a powerful tool to probe the formation of the
above-mentioned addition product, because the yC¼O band
of the aldehyde is strong and symmetrical. Therefore, the
area of this peak has been followed as a function of time;
the results are shown in Figure 3.
Aspart(A)ofFigure3clearlyshows,thereispracticallyno
change in the area of the yC¼O band of the benzaldehyde/
AlMeImCl mixture as a function of time. This indicates that
the IL-Cl is stable under these experimental conditions.
On the other hand, the peak area for the corresponding
AlMeImF (part B) decreased to 38% of its original value after
5 h and changed very little afterword; most probably due to
the reversibility of above-indicated (ylide-benzaldehyde)
addition reaction. The results of part (C) will be discussed
later.
The occurrence of side reactions can be explained on
the bases of abstraction of the relatively acidic C2-H
of the imidazolium moiety by the fluoride ion (Fꢁ is
isoelectronic with OHꢁ).[30–32,51] In order to test our
assumption (hydrogen atom abstraction), evaporation of
methanol from AlMeImF solution (no DMSO added) has
been carried out in the presence of added benzaldehyde.
For comparison, the same experiment was repeated with
AlMeImCl. The idea of this experiment is to ‘‘trap’’ the
intermediate ylide formed by abstraction of C2-H of
the imidazolium ring as a stable addition product to
benzaldehyde, as shown in Scheme 3. This approach is
similar to that employed for probing the effect of a base
(3-hydroxyquinuclidine) on 1-butyl-3-methylimidazolium
chloride, where the aldehyde CH¼O peak at 10.0 ppm has
been replaced by a peak at 6.35 ppm.[30]
1
In summary, H NMR and FTIR results of the trapping
experiment corroborate that the side reaction in case of
AlMeImF is F–-mediated hydrogen abstraction from the
substituted imidazolium ring.
3.3. Solution of the Problem of Side Reactions
The following question now arises: can the eliminationside
reaction be stopped/suppressed by addition of DMSO
before methanol evaporation? Our reasoning rests on the
ability of DMSO to associate with ILs whose counter-ions
contains fluorine, e.g., the 1-butyl-3-methylimidazolium-
First, we discuss the experiment of AlMeImCl. As can be
seen from Figure 1, the 1H NMR spectrum obtained for the
AlMeImCl/benzaldehyde mixture after methanol evapora-
tion is simply the sum of the peaks of both components. In
and
1-butyl-2,3-dimethylimidazolium
tetrafluorobo-
rate.[52] We have reasoned that DMSO is especially suited
for the job, because: (i) solvation of (Fꢁ), if it occurs, should
Macromol. Biosci. 2013, 13, 191–202
ß 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
196
www.MaterialsViews.com

Successful Application of an Ionic Liquid Carrying the Fluoride Counter-ion . . .
www.mbs-journal.de
Figure 1. 1H NMR spectrum for the mixture of AlMeImCl and benzaldehyde after a two-stage methanol evaporation (200–10 mm Hg, 50 8C;
2 mmHg, for ca. 4 h, at 60 8C). The region 6.50–5.80 ppm is expanded in the inset; it refers to the same mixture after the first stage of
methanol evaporation. There is no peak at ca. 6.3 ppm, attributed to the IL-Cl derived ylide-aldehyde addition product.
decrease its basicity, hence stop/suppress the side reac-
tions; (ii) being a strongly dipolar solvent, DMSO should
form hydrogen-bond with the C2-H of the imidazolium
ring, hence decrease its interaction with the (Fꢁ); (iii) the
concomitant removal of DMSO during methanol evapora-
tion should be small, because of its high b.p., 189 8C; (iv)
cellulose dissolution in the resulting IL-F/DMSO should be
favorable because of the high capacity of the molecular
solvent in causing biopolymer swelling (DMSO ranks
second only to water).[53,54]
and benzaldehyde mixture is smaller than that shown in
Figure 2 (for AlMeImF/benzaldehyde). Therefore, the
addition of DMSO leads to suppression of H-abstraction
reaction to a large extent, probably due to the AlMeImF-
DMSO interactions, as depicted in Scheme 4.[21,52]
The rationale for these structures will be discussed in
Section 3.6; vi.
A clear indication that our expectation is valid can be
reached by comparing the DS results analyzed for
(commercial) cellulose triethanoate after treatment
(80 8C; 30 W; 3 h) with ILs, point above (Section 3.2; ii):
Although part (C) of Figure S1 of Supporting Information
shows that AlMeImF/DMSO is still dark-colored, its 1H NMR
spectrum (Figure S4 of Supporting Information) shows
fewer additional peaks; no gas evolution has been observed
during methanol evaporation, provided that the mole
fraction of DMSO in the mixture, xDMSO is ꢄ0.5. Part (C) of
Figure 3 shows that the extent of ylide trapping has
decreased noticeably after 5 h, from 62 to 20%, when DMSO
has been added before alcohol evaporation. Additionally,
original sample, 2.70; AlMeImCl, 2.60; AlMeImCl0.22
/
DMSO0.78, 2.65; AlMeImF, 1.1; AlMeImF0.22/DMSO0.78, 2.5.
That is, addition of DMSO also suppressed fluoride ion-
mediated ester hydrolysis.
It is legitimate to suggest that the ion-exchange and
neutralization steps, vide Experimental, could have been
carried out in neat DMSO instead of methanol. To be
successful/efficient, ion-exchange reactions are usually
carried out in the laboratory by using relatively dilute
solutions. Therefore, a large volume of DMSO would have
1
the intensity of the singlet at ca. 6.3 ppm in the H NMR
spectrum (not shown) obtained for the AlMeImF/DMSO
Macromol. Biosci. 2013, 13, 191–202
ß 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
197
www.MaterialsViews.com

R. Casarano, O. A. El Seoud
www.mbs-journal.de
Figure 2. 1H NMR spectrum for the mixture of AlMeImF and benzaldehyde after a two-stage methanol evaporation (200–10 mm Hg, 50 8C;
2 mmHg, for ca. 4 h, at 60 8C). The region 6.50–5.80 ppm is expanded in the inset; it refers to the same mixture after the first stage of
methanol evaporation.
been needed; this excess should be partially evaporated in
order to obtain IL-F/DMSO of the required composition.
This procedure, however, would have been both energy
intensive and expensive, in view of the large difference in
the b.p. (65; 189 8C) and relative cost (1 and 3.3; ACS-grade
solvents; prices for one liter) of methanol and DMSO,
respectively.
AlMeImF/DMSO has also been successfully employed for
the synthesis of other esters, butanoate and hexanoate, see
Table 2. The dependence of DS on the anhydride employed
is similar to that observed elsewhere, i.e., ethanoic ꢅ
hexanoic > butanoic.[19,55] As recently shown, this depen-
dence of DS on the length of the acyl moiety of the
anhydride is due to subtle changes in- and compensations
of the enthalpy and entropy of activation.[56]
Acetylation of (fibrous) eucalyptus cellulose has also
been successfully carried out in AlMeImF/DMSO, as shown
in Table 2. The DS values of the acetates increase with
increasingxDMSO inthemixture, similartothatobservedfor
the acetylation of MCC. As expected, the DS value obtained
for MCC is greater than that for eucalyptus (Table 2), due to
differences in DPv and, most certainly, in the accessibility of
microcrystalline- and fibrous biopolymers.[57] As observed
3.4. Acylation of Celluloses in AlMeImF/DMSO and
Comparison with the Reactions in AlMeImCl and
AlMeImCl/DMSO
Acetylation of MCC was carried out in AlMeImF/DMSO as
solvent; the results are shown in Figure 4 and Table 2.
The DS values increase as a function of increasing xDMSO
reaching 2.9 at AlMeImF0.22/DMSO0.78, which is equal to
that of the reaction in neat AlMeImCl (2.95, Table 1), using
onlyonethirdinmassoftheIL.Despitethedarkambercolor
of AlMeImF/DMSO solution, the color of cellulose acetate
obtained is similar to that of the reaction in AlMeImCl
(Figure S1 of Supporting Information).
for MCC, the acetylation of fibrous cellulose in AlMeImF0.22
/
DMSO0.78 gave DS value equal to that in neat AlMeImCl.
Finally, the efficiency of AlMeImF/DMSO relative to
the corresponding AlMeImCl/DMSO is shown by the large
difference in DS at the same xDMSO, Table 2.
Macromol. Biosci. 2013, 13, 191–202
ß 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
198
www.MaterialsViews.com

Successful Application of an Ionic Liquid Carrying the Fluoride Counter-ion . . .
www.mbs-journal.de
1.0
"Zero"- area = 1.00
5 h - area = 0.99
A
0.8
0.6
0.4
0.2
0.0
24 h - area = 0.99
48 h - area = 0.98
72 h - area = 0.97
Scheme 4. Representative ‘‘snap-shots’’ of the proposed inter-
actions of AlMeImF with DMSO. Structure (a) shows one possib-
ility for a six-member ‘‘ring’’ solvation of (Fꢁ), involving DMSO,
the positively charged nitrogen-, and C2-H of the imidazolium
ring. The same ring size has been maintained in (b), but the
stoichiometry DMSO/IL-F is 2:1. Structure (c) shows the cellulose/
IL-F/DMSO interaction via an 8-member ring formation.
1800 1750 1700 1650 1600 1550 1500
Wavenumber (cm-1)
1.0
"Zero" - area = 1.00
5 h - area = 0.38
24 h - area = 0.37
48 h - area = 0.35
72 h - area = 0.35
B
0.8
0.6
0.4
0.2
0.0
1800 1750 1700 1650 1600 1550 1500
Wavenumber (cm-1)
1.0
"Zero" - area = 1.00
5 h - area = 0.80
24 h - area = 0.72
48 h - area = 0.68
72 h - area = 0.64
C
0.8
0.6
0.4
0.2
0.0
Figure 4. Dependence of the DS of cellulose acetate on the mole
fraction of DMSO in the mixture of AlMeImF/DMSO. Experimen-
tal conditions: 3 h, 80 8C, 30 W, and Ac2O/AGU molar ratio of 6;
data for MCC.
1800 1750 1700 1650 1600 1550 1500
Wavenumber (cm-1)
Figure 3. The 1 800–1 500 cmꢁ1 region of the IR spectra of 1:1 molar
mixtures of benzaldehyde and (A) AlMeImCl, (B) AlMeImF, (C)
AlMeImF0.22/DMSO0.78 acquired after ‘‘zero’’, 5, 24, 48, and 72 h.
anhydride, as shown by the reaction:
RCOOCOR þ AlMeImF ! RCOF þ AlMeImO2CR
Expanded region from 1 800 to 1 500 cmꢁ1
.
The above-raised question is based on the following
facts: (Fꢁ) is a powerful nucleophile, especially when it is
desolvated;[61] the formation of acyl fluorides has been
experimentally demonstrated in, e.g.: the preparation of
perhalo-alkyl acrylate and methacrylate esters;[62] the
hydrolysis of acetic anhydride in aqueous medium;[63]
the reaction of acetyl chloride with TBAF/acetic acid;[64]
the reaction of TAAF with carboxylic acid anhydrides in
DMSO.[40]
3.5. Nature of the Acylating Agent: Detection of Acyl
Fluoride Formation by FTIR
It is now recognized that as solvents, ILs are not always
‘‘spectators’’; they may participate in the reaction.[58–60]
This participation, in fact, is the base of the so-called task-
specific ILs.[16] Is it possible that a fraction of the ester is
produced via the reaction of cellulose with acyl fluoride,
RCOF? In principle, the latter can be formed by the
nucleophilic attack of the fluoride counter-ion on the
We have employed FTIR in order to examine solutions of
acid anhydrides(ethanoic- or hexanoic) in AlMeImF/DMSO,
in absence (spectra not shown) and presence of cyclohex-
ylmethanol, CHM. The latter carries a primary (OH) group
and has been successfully employed as a model for the
Macromol. Biosci. 2013, 13, 191–202
ß 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
199
www.MaterialsViews.com

R. Casarano, O. A. El Seoud
www.mbs-journal.de
Table 2. Results of acylation of MCC, or eucalyptus cellulose, in AlMeImF, AlMeImF/DMSO, AlMeImCl, and AlMeImCl/DMSO.a)
Solvent
Cellulose
Carboxylic anhydride
Ester DS
b),c)
b),c)
b),c)
b),c)
c)
AlMeImF0.74/DMSO0.26
AlMeImF0.53/DMSO0.47
AlMeImF0.33/DMSO0.67
AlMeImF0.22/DMSO0.78
AlMeImF0.33/DMSO0.67
AlMeImF0.33/DMSO0.67
AlMeImF0.53/DMSO0.47
AlMeImF0.33/DMSO0.67
AlMeImF0.22/DMSO0.78
AlMeImCl
MCC
Ethanoic
Ethanoic
Ethanoic
Ethanoic
Butanoic
Hexanoic
Ethanoic
Ethanoic
Ethanoic
Ethanoic
Ethanoic
Ethanoic
0.4
1.55
2.5
2.9
2.1
2.65
1.1
1.7
2.1
2.2
1.2
1.1
MCC
MCC
MCC
MCC
c)
MCC
c)
Eucalyptus
Eucalyptus
Eucalyptus
Eucalyptus
MCC
c)
c)
c)
AlMeImCl0.22/DMSO0.78
AlMeImCl0.22/DMSO0.78
c)
Eucalyptus
a)Experimental conditions: 3 h, 80 8C, 30 W, and carboxylic anhydride/AGU molar ratio of 6. Eucalyptus cellulose was mercerized;
b)these data were plotted in Figure 4; c)the following are the weight- and mole-fractions employed for AlMeImF/DMSO: 0.84/0.16, 0.74/
0.26; 0.67/0.33, 0.53/0.47; 0.47/0.53, 0.33/0.67; 0.34/0.66, 0.22/0.78. The corresponding values for AlMeImCl/DMSO are: 0.36/0.64, 0.22/
0.78, respectively.
C6-OH of the AGU.[56] Room temperature has been
100
employed because of the volatility of CH3COF, if it is
formed, b.p. ¼ 20 8C.[40,65]
80
1822
As expected, the anhydrides show the characteristic
asymmetric and symmetric C¼O stretching vibrations at
60
1 822, 1 742, and 1 813, 1 742 cmꢁ1, for ethanoic- and
1842
40
20
0
hexanoic anhydride, respectively. The formation of acyl
fluorides is observed right after mixing at 1 842 cmꢁ1
(Figure 5A, CH3COF), and 1 837 cmꢁ1 (Figure 5B, C5H11COF),
in agreement with literature values.[66] As shown in
Figure 5B, as a function of time, the intensity of nC¼O of
1721
"Zero"
1600 1500
A
2000
1900
1800
1700
RCOF decreased, andthatof nC¼O ofC5H11CO2H(1722 cmꢁ1
)
Wavenumber (cm-1)
increased, probably due to the reaction with CHM
and/or with adventitious water. Although it is known
that acyl fluorides have lower reactivity, relative to
other acyl halides, e.g., in acyl-transfer-,[67,68] and Friedel-
Crafts reactions,[69] it is safe to assume that at least a
part of the product is formed by reaction with RCOF.
100
80
60
40
20
0
1813
1837
3.6. A Tentative Explanation of the Role of DMSO
"Zero"
5 h
24 h
1722
We offer tentative explanation for the suppression of the
side reactions, and for the higherefficiency of IL-F/DMSO, as
compared to its IL-Cl/DMSO counterpart. Efficient solvation
of (Fꢁ) of AlMeImF is expected to decrease its basicity, hence
its tendency for ester hydrolysis and imidazolium C2-H
elimination. Structures (a) and (b) of Scheme 4 are derived
from the observation that DS increases noticeably
at ca. xDMSO ꢄ 0.5, corresponding to 1:1 IL-F-DMSO complex.
B
2000
1900
1800
1700
1600
1500
Wavenumber (cm-1)
Figure 5. The 2 000–1 500 cmꢁ1 IR spectral region of the mixtures:
(AlMeImF0.33/DMSO0.67); ethanoic anhydride, and CHM right after
mixing (A); AlMeImF0.33/DMSO0.67; hexanoic anhydride, and CHM
(B) as a function of time.
Macromol. Biosci. 2013, 13, 191–202
ß 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
200
www.MaterialsViews.com

Successful Application of an Ionic Liquid Carrying the Fluoride Counter-ion . . .
www.mbs-journal.de
This solvation may occur via the formation of the relatively
stable six-membered ‘‘ring’’, containing one-, or two DMSO
molecules. Both show the hydrogen bonds formed between
C2-HꢆꢆꢆO ¼ S, andC2-HꢆꢆꢆFꢁ, inagreementwithpublished
data on IL-BF4.[52,70,71] As shown elsewhere,[72] DMSO
does not perturb noticeably the interactions between the
ions of IL-Cl, e.g., 1-butyl-3-methylimidazolium chloride,
BuMeImCl, so that ILs clusters persist, even at low
concentration of the IL-Cl in this molecular solvent. This
persistencemaybededuced, indirectly, fromthereportthat
up to 15 wt.-% ofDMSO hasno apparent effecton the ability
of BuMeImCl to dissolve polysaccharide-rich biomass.[72]
This may be taken to indicate that the structure of this IL-Cl
is not largely perturbed by DMSO. In fact, a clear solution of
extensively swollen MCC in BuMeImCl/DMSO, xDMSO 0.92,
was obtained by Rinaldi.[73] In summary, the IL-F interacts
more strongly with the strongly dipolar DMSO than the
corresponding IL-Cl, this leads to efficient solvation of
strongly basic fluoride ion, and suppression of its capacity
to induce the above-mentioned elimination reaction, and
(general-base catalyzed) hydrolysis of the ester formed. Our
results can be analyzed quantitatively in terms of the
dependence of DS on, e.g., solvent basicity, dipolarity, and
polarizability, as measured by solvatochromic probes.[20]
This analysis, however, is not feasible at this stage; DS for
the reaction in IL-F dissolved a group of dipolar aprotic
solvents are necessary in order to obtain statistically
significant correlations.
the imidazolium moiety. These side reactions have
precluded the use of IL-Fs as media for chemical reactions,
including the derivatization of cellulose and other biopo-
lymers. In order to solve this problem, we have taken
advantage of the relatively strong solvation of (Fꢁ) by
DMSO. This reduce C2-H abstraction by a factor of 2.1 (5 h,
room temperature), turning acylation of cellulose feasible.
It also suppresses general-base catalyzed ester hydrolysis.
Equally important, however, is the fact the medium
employed, AlMeImF0.22/DMSO0.78 is composed essentially
of (relatively inexpensive) DMSO; the IL-F is acting as an
electrolyte, akin to LiCl (in DMAC) and R4NF.xhydrate (in
DMSO). Therefore, the extra labor and cost involved in
obtaining AlMeImF/DMSO is more than compensated for
by using the IL-F as an electrolyte, not a solvent. These
results make AlMeImF/DMSO a potential candidate for the
dissolution/derivatization of other biopolymers, and for
large-scale processing of biomass.
Acknowledgements: We thank the FAPESP (State of S˜ao Paulo
Research Foundation) for financial support and a postdoctoral
fellowship to R. Casarano; the CNPq (National Council for Scientific
and Technological Research) for a productivity fellowship to O. A.
El Seoud. We also thank Paulo A. R. Pires for the NMR analyses and
Cezar Guizarro for his help. This work has been carried out within
the frame of INCT-Cata´lise.
With regard to the higher efficiency of AlMeImF/DMSO,
consider structure (c) of Scheme 4, that is derived from
structure (b) by insertion of a cellulose molecule, leading
to the formation of an 8-membered ‘‘ring’’. Structure (c),
if it is formed, is expected to enhance not only the
accessibility of cellulose, but also its reactivity due to
partial weakening of Cell-O-H, with concomitant
increase in the nucleophilicity of Cell-O-H. A possible
support for structure (c) comes from the observation that
the biopolymer solution in AlMeImF/DMSO turns clear
only after addition of the acid anhydride (vide
Experimental), i.e., when such cyclic, cellulose-containing
structures may have started to become more soluble
due to the acylation reaction proper. Because of the
weaker IL-Cl-DMSO interactions, the formation of structure
(c) may be less favorable, leading to the observed lower DS
for cellulose acylation in AlMeImCl/DMSO, relative to
the IL-F counterpart.
Received: August 11, 2012; Revised: October 9, 2012; Published
online: January 9, 2013; DOI: 10.1002/mabi.201200296
Keywords: acyl fluorides; 1-allyl-3-methylimidazolium fluoride;
biopolymer-(ionic liquid)-DMSO interactions; cellulose esterifica-
tion
[1] M. Kim, D. F. Day, J. Ind. Microbiol. Biotechnol. 2011, 38, 803.
[2] C. Cook, A. J. Devoto, Sci. Food Agric. 2011, 91, 1729.
[3] D. R. Dodds, R. A. Gross, Science 2007, 318, 1250.
[4] A. C. O’Sullivan, Cellulose 1997, 4, 173.
[5] D. Klemm, B. Phillip, T. Heinze, U. Heinze, W. Wagenknecht,
Comprehensive Cellulose Chemistry, Vol. 1, Wiley-VCH, Wein-
heim 1998.
[6] A. Pinkert, K. N. Marsh, S. Pang, M. P. Staiger, Chem. Rev. 2009,
109, 6712.
[7] T. R. Dawsey, C. L. McCormick, J. Macromol. Sci. Rev. Macromol.
Chem. Phys. 1990, C30, 405.
4. Conclusion
[8] P. Burkat, Polym. News 1990, 15, 170.
[9] T. Heinze, W. G. Glasser, ACS Symp. Ser. 1998, 688, 2.
[10] O. A. El Seoud, G. A. Marson, G. T. Ciacco, E. Frollini, Macromol.
Chem. Phys. 2000, 201, 882.
The use of AlMeImF for cellulose dissolution and subse-
quent derivatization is problematic due to the occurrence
of Fꢁ-mediated side reactions. These include cellulose ester
hydrolysis and abstraction of the relatively acidic C2-H of
[11] T. Heinze, R. Dicke, A. Koschella, A. H. Kull, W. Koch, Macro-
mol. Chem. Phys. 2000, 201, 627.
[12] O. A. El Seoud, T. Heinze, Adv. Polym. Sci. 2005, 186, 103.
Macromol. Biosci. 2013, 13, 191–202
ß 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
201
www.MaterialsViews.com

R. Casarano, O. A. El Seoud
www.mbs-journal.de
[13] M. E. Zakrzewska, E. Bogel-Lukasik, R. Bogel-Lukasik, Energy
Fuels 2010, 24, 737.
[43] ASTM D 871-96 Standard Test Methods of Testing Cellulose
Acetate (Solution Method; Procedure A) 2004.
[44] ASTM D 1795-94 Standard Test Methods for Intrinsic
[14] T. Liebert, ACS Symp. Ser. 2010, 1033, 3.
[15] J. P. Hallett, T. Welton, Chem. Rev. 2011, 111, 3508.
[16] J. H. Davis, Jr., Chem. Lett. 2004, 33, 1072.
[17] O. A. El Seoud, A. Koschella, L. C. Fidale, S. Dorn, T. Heinze,
Biomacromolecules 2007, 8, 2629.
Viscosity of Cellulose, 2001.
[45] Q. Xu, Q. Wang, L. Liu, J. Appl. Polym. Sci. 2008, 107, 2704.
[46] Y. Deng, T. Long, D. Zhang, J. Chen, S. Gan, J. Chem. Eng. Data
2009, 54, 2470.
[18] J. Vitz, T. Erdmenger, C. Haensch, U. S. Schubert, Green Chem.
2009, 11, 417.
[47] W. Xiao, Q. Chen, Y. Wu, T. Wu, L. Dai, Carbohydr. Polym. 2011,
83, 233.
[19] S. Possidonio, L. C. Fidale, O. A. El Seoud, J. Polym. Sci., Part A:
Polym. Chem. 2010, 48, 134.
[48] R. Casarano, O. A. El Seoud, Patent pending.
[49] D. Xu, K. J. Edgar, Biomacromolecules 2012, 13, 299.
[50] T. Rosenau, A. Potthast, I. Adorjan, A. Hofinger, H. Sixta,
H. Firgo, P. Kosma, Cellulose 2002, 9, 283.
[20] C. Reichardt, T. Welton, Solvents and Solvent Effects in Organic
Chemistry, Wiley-VCH, Weinheim 2011, pp. 509, 425.
[21] A. Pinkert, K. N. Marsh, S. Pang, Ind. Eng. Chem. Res. 2010, 49,
11121.
[51] S. Kluge, J. Weston, Biochemistry 2005, 44, 4877.
[52] J.-C. Jiang, K.-H. Lin, S.-C. Li, P.-M. Shih, K.-C. Hung, S. H. Lin,
H.-C. Chang, J. Chem. Phys. 2011, 134, 044506/1.
[53] O. A. El Seoud, L. C. Fidale, N. Ruiz, M. L. O. D’Almeida,
E. Frollini, Cellulose 2008, 15, 371.
[22] B. Lindman, G. Karlstroem, L. Stigsson, J. Mol. Liq. 2010,
156, 76.
[23] R. P. Swatloski, S. K. Spear, J. D. Holbrey, R. D. Rogers, J. Am.
Chem. Soc. 2002, 124, 4974.
[54] L. C. Fidale, N. Ruiz, T. Heinze, O. A. El Seoud, Macromol. Chem.
Phys. 2008, 209, 1240.
[24] H. Zhao, G. A. Baker, Z. Song, O. Olubajo, T. Crittle, D. Peters,
Green Chem. 2008, 10, 696.
[55] T. Heinze, S. Dorn, M. Schoebitz, T. Liebert, S. Koehler,
F. Meister, Macromol. Symp. 2008, 262, 8.
[25] Q.-L. Kuang, J.-C. Zhao, Y.-H. Niu, J. Zhang, Z.-G. Wang, J. Phys.
Chem. B 2008, 112, 10234.
[56] H. Nawaz, R. Casarano, O. A. El Seoud, Cellulose 2012, 19,
199.
[26] P. Maeki-Arvela, I. Anugwom, P. Virtanen, R. Sjo¨holm, J. P.
Mikkola, Ind. Crop. Prod. 2010, 32, 175.
[57] L. A. Ramos, D. L. Morgado, O. A. El Seoud, V. C. da Silva,
E. Frollini, Cellulose 2011, 18, 385.
[27] O. A. El Seoud, V. C. da Silva, S. Possidonio, R. Casarano, E. P. G.
Areˆas, P. Gimenes, Macromol. Chem. Phys. 2011, 212, 2541.
[28] C. Rijksen, R. D. Rogers, J. Org. Chem. 2008, 73, 5582.
[29] A. Maiti, P. F. Pagoria, A. E. Gash, T. Y. Han, C. A. Orme, R. H.
Gee, L. E. Fried, Phys. Chem. Chem. Phys. 2008, 10, 5050.
[30] V. K. Aggarwal, I. Emme, A. Mereu, Chem. Commun. 2002,
1612.
[58] R. P. Swatloski, J. D. Holbrey, R. D. Rogers, Green Chem. 2003, 5,
361.
[59] S. Chowdhury, R. S. Mohan, J. L. Scott, Tetrahedron 2007, 63,
2363.
[60] M. Schrems, G. Ebner, F. Liebner, E. Becker, A. Potthast,
T. Rosenau, ACS Symp. Ser. 2010, 1033, 149.
[31] S. T. Handy, M. J. Okello, J. Org. Chem. 2005, 70, 1915.
[32] E. Ennis, S. T. Handy, Curr. Org. Synth. 2007, 4, 381.
[33] G. Ebner, S. Schiehser, A. Potthast, T. Rosenau, Tetrahedron
Lett. 2008, 49, 7322.
[61] V. M. Vlasov, J. Fluorine Chem. 1993, 61, 193.
[62] A. G. Pittman, D. L. Sharp, R. E. Lundin, J. Polym. Sci., Part A:
Polym. Chem. 1966, 10, 2637.
[63] C. A. Bunton, J. H. Fendler, J. Org. Chem. 1967, 32, 1547.
[64] J. Emsley, V. Gold, F. Hibbert, W. T. A. Szeto, J. Chem. Soc.,
Perkin Trans. 2: Phys. Org. Chem. 1988, 6, 923.
[65] G. Olah, I. Kuhn, J. Org. Chem. 1961, 26, 237.
[66] J. A. Ramsey, J. A. Ladd, J. Chem. Soc. B: Phys. Org. 1968, 118.
[67] G. Swain, C. B. Scott, J. Am. Chem. Soc. 1953, 75, 141.
[68] A. Kivinen, ‘‘Mechanisms of Substitution at the COX Group’’,
in The Chemistry of Acyl Halides, S. Patai, Ed., J. Wiley & Sons,
New York 1972, pp. 177–230.
[34] S. Sowmiah, V. Srinivasadesikan, M.-C. Tseng, Y.-H. Chu,
Molecules 2009, 14, 3780.
[35] R. K. Sharma, J. L. Fry, J. Org. Chem. 1983, 48, 2112.
[36] M. Makosza, A. A. Chesnokov, Tetrahedron 2003, 59, 1995.
[37] E. Ennis, S. T. Handy, Molecules 2009, 14, 2235.
[38] S. Chempath, J. M. Boncella, L. R. Pratt, N. Henson, B. S. Pivovar,
J. Phys. Chem. C 2010, 114, 11977.
[39] M. Ueki, K. Kai, M. Amemiya, H. Horino, H. Oyamada, J. Chem.
Soc., Chem. Commun. 1988, 414.
[69] Y. Yamasi, Bull. Chem. Soc. Jpn. 1961, 34, 480.
[70] A. Mele, C. D. Tran, S. H. D. Lacerda, Angew. Chem., Int. Ed.
2003, 42, 4364.
[40] R. Casarano, H. Nawaz, S. Possidonio, V. C. da Silva, O. A.
El Seoud, Carbohydr. Polym. 2011, 86, 1395.
[41] W. L. F. Armagero, C. L. L. Chai, Purification of Laboratory
Chemicals, Elsevier, New York 2003.
[71] L. Zhang, Y. Wang, Z. Xu, H. Li, J. Phys. Chem. B 2009, 113, 5978.
[72] R. C. Remsing, Z. Liu, I. Sergeyev, G. Moyna, J. Phys. Chem. B
2008, 112, 7363.
[42] R. Casarano, L. C. Fidale, C. M. Lucheti, T. Heinze, O. A. El Seoud,
Carbohydr. Polym. 2011, 83, 1285.
[73] R. Rinaldi, Chem. Commun. 2011, 47, 511.
Macromol. Biosci. 2013, 13, 191–202
ß 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
202
www.MaterialsViews.com