molecules
Article
Nano-Structural Investigation on Cellulose Highly
Dissolved in Ionic Liquid: A Small Angle X-ray
Scattering Study
Takatsugu Endo 1, *, Shota Hosomi 1 , Shunsuke Fujii 1 , Kazuaki Ninomiya 2 and
Kenji Takahashi 1, *
1
2
*
Faculty of Natural System, Institute of Science and Engineering, Kanazawa University, Kakuma-machi,
Kanazawa 920-1192, Japan; [email protected] (S.H.); [email protected] (S.F.)
Institute for Frontier Science Initiative, Kanazawa University, Kakuma-machi, Kanazawa 920-1192, Japan;
[email protected]
Correspondence: [email protected] (T.E.); [email protected] (K.T.);
Tel.: +81-76-234-4807 (T.E.); +81-76-234-4828 (K.T.)
Academic Editor: Hua Zhao
Received: 21 December 2016; Accepted: 18 January 2017; Published: 21 January 2017
Abstract: We investigated nano-structural changes of cellulose dissolved in 1-ethyl-3methylimidazolium acetate—an ionic liquid (IL)—using a small angle X-ray scattering (SAXS)
technique over the entire concentration range (0–100 mol %). Fibril structures of cellulose disappeared
at 40 mol % of cellulose, which is a significantly higher concentration than the maximum concentration
of dissolution (24–28 mol %) previously determined in this IL. This behavior is explained by the
presence of the anion bridging, whereby an anion prefers to interact with multiple OH groups of
different cellulose molecules at high concentrations, discovered in our recent work. Furthermore, we
observed the emergence of two aggregated nano-structures in the concentration range of 30–80 mol
%. The diameter of one structure was 12–20 nm, dependent on concentration, which is ascribed to
cellulose chain entanglement. In contrast, the other with 4.1 nm diameter exhibited concentration
independence and is reminiscent of a cellulose microfibril, reflecting the occurrence of nanofibrillation.
These results contribute to an understanding of the dissolution mechanism of cellulose in ILs. Finally,
we unexpectedly proposed a novel cellulose/IL composite: the cellulose/IL mixtures of 30–50 mol %
that possess liquid crystallinity are sufficiently hard to be moldable.
Keywords: ionic liquids; cellulose; small angle X-ray scattering; dissolution; composites
1. Introduction
Cellulose—the major component of wood—is renewable, and is the most abundant biopolymer
on Earth; therefore, its advanced unitization is crucial for a sustainable society. However, the severe
limiting factor is that cellulose does not dissolve in any conventional solvents. A new class of
cellulose solvents—ionic liquids (ILs)—was discovered in 2002 [1]. ILs are defined as salts that
are liquid around room temperature, and some of them are able to efficiently dissolve cellulose [2,3].
ILs have some outstanding properties as cellulose solvents compared with those used previously:
high designability (any appropriate physicochemical properties can be introduced by modifying
the cation/anion structures or altering their combination), single component, practically nonvolatile
and nonflammable, and highly thermally/chemically/electrochemically stable. Since these green
solvents can be a breakthrough for cellulose science, understanding how cellulose dissolves in ILs is
a very important issue. Many studies have been conducted thus far to contribute to this aim [2–5].
The most crucial finding is that IL anions play a major role in the dissolution by disrupting the
Molecules 2017, 22, 178; doi:10.3390/molecules22010178
www.mdpi.com/journal/molecules
Molecules 2017, 22, 178
2 of 9
inter- and intramolecular hydrogen-bonding network of cellulose. These anions mainly interact with
the hydrogen of the OH groups in a 1:1 ratio [6], although it was suggested by molecular dynamics
(MD) simulations that some anions participate in multiple interactions with the OH groups [7,8].
The role of the cations seems to be less significant; nevertheless, they unquestionably have some
Molecules 2017, 22, 178
2 of 9
impact on the dissolution. Although the definitive role is still controversial, MD simulations have
intramolecular
network
of cellulose. between
These anions
with
the
indicated the
presence hydrogen-bonding
of weak hydrophobic
interactions
the mainly
cationsinteract
and the
glucose
ring of
hydrogen
of the
OH groups
in aaromatic
1:1 ratio [6],cations
althoughtend
it wastosuggested
molecular
dynamics
(MD) [2,3].
cellulose [4,5],
which
explains
why
exhibit by
high
dissolution
ability
simulations that some anions participate in multiple interactions with the OH groups [7,8]. The role
It is well
known that the concentration of a solution governs the state of dissolution of solute
of the cations seems to be less significant; nevertheless, they unquestionably have some impact on
molecules.theParticularly
at high
(e.g.,
lyotropic
liquid
dissolution. Although
the concentrations,
definitive role is still self-assembly
controversial, MDphenomena
simulations have
indicated
the
weak hydrophobic
interactions However,
between the cations
and the glucose
ring of cellulose
crystal or presence
micelleofformation)
are observed.
a concentration
regime
of more[4,5],
than 30 mol
why as
aromatic
cations
tend to exhibit high
ability [2,3].
% (mol % which
here explains
is defined
glucose
unit/(glucose
unitdissolution
+ IL))—which
is considered to exceed the
It is well known that the concentration of a solution governs the state of dissolution of solute
maximummolecules.
solubilityParticularly
because at
the
ratio
of
IL
to
the
OH
groups
of
cellulose
is
greater than 1:1—has been
high concentrations, self-assembly phenomena (e.g., lyotropic liquid crystal
completely
overlooked.
Recently,
we
performed
a
structural
investigation
on
in
or micelle formation) are observed. However, a concentration regime of more than
30cellulose
mol % (moldissolved
%
here is defined as glucoseacetate
unit/(glucose
unit + IL))—which is
considered
to exceed the maximum
1-ethyl-3-methylimidazolium
([Emim][OAc])—as
the
most representative
IL for cellulose
solubility because
the ratio
of IL to the OH
groups
of cellulose
is greater
than 1:1—has
been
dissolution—over
the entire
concentration
range
(0–100
mol %)
by means
of wide-angle
X-ray
completely overlooked. Recently, we performed a structural investigation
on
cellulose
dissolved
in
13
scattering 1-ethyl-3-methylimidazolium
with the aid of quantumacetate
chemical
calculations and C solid-state NMR spectroscopy [9].
([Emim][OAc])—as the most representative IL for cellulose
We demonstrated
that the
concentrations
≥30 mol
% showed
dissolution—over
thestates
entire at
concentration
range (0–100
mol %)
by means anomalous
of wide-angle deconstruction
X-ray
scattering with
the aid of quantum
chemical calculations
and 13C solid-state
NMR
and restructuring
of cellulose,
as schematically
summarized
in Figure
1. spectroscopy
Even at 40[9].
mol %, the
We
demonstrated
that
the
states
at
concentrations
≥30
mol
%
showed
anomalous
deconstruction
and
original crystalline structure of cellulose is absent. Furthermore, a new cellulose/IL complex with high
restructuring of cellulose, as schematically summarized in Figure 1. Even at 40 mol %, the original
structural crystalline
periodicity
emerges at approximately 35 mol %. It was revealed that these characteristic
structure of cellulose is absent. Furthermore, a new cellulose/IL complex with high
features originated
from theemerges
anionatbridging
phenomenon,
is when
one characteristic
anion interacts with
structural periodicity
approximately
35 mol %. It which
was revealed
that these
features
originated
from thecellulose
anion bridging
phenomenon,
which is when
anion
with
multiple OH
groups
of different
chains.
At concentrations
≤25one
mol
%, interacts
the anion/OH
group
OH groups
of different cellulose
chains. At concentrations
≤25 mol
the anion/OH
group bridging
interactionmultiple
is mainly
1:1 (non-bridging),
as previously
reported [6–8].
In%,contrast,
the anion
interaction is mainly 1:1 (non-bridging), as previously reported [6–8]. In contrast, the anion bridging
state is energetically
preferred at the higher concentrations, which induces intriguing structural changes
state is energetically preferred at the higher concentrations, which induces intriguing structural
in cellulose.
changes in cellulose.
Figure 1. Schematic summary of Reference [9]. Below 25 mol %, the anions mainly interact with the
Figure 1. Schematic
Reference
[9]. Below
25 some
mol %,
the of
anions
mainly
interact with the
OH groups ofsummary
cellulose inof
a 1:1
ratio (non-bridging),
while
portion
the anions
have multiple
interactions
with in
theaOH
In the concentration
range
of 25–40
mol %,
state multiple
OH groups
of cellulose
1:1groups
ratio [7,8].
(non-bridging),
while
some
portion
ofthe
theinteraction
anions have
of with
the anion-OH
group gradually
from non-bridging
to 25–40
bridging,
induces
the
interactions
the OH groups
[7,8]. In switches
the concentration
range of
molwhich
%, the
interaction
state of
emergence of the cellulose/[Emim][OAc] (1-ethyl-3-methylimidazolium acetate) complex structure.
the anion-OH group gradually switches from non-bridging to bridging, which induces the emergence
The anions bridge cellulose chains, forming a sheet structure. The sheets are stacked upon each other,
of the cellulose/[Emim][OAc]
(1-ethyl-3-methylimidazolium
acetate)
structure.
The anions
intercalated by the cations,
forming a layered structure. Note that
the exactcomplex
structure is
less ordered
bridge cellulose
forming
a sheet chains
structure.
The
sheets are sheets
stacked
each other,
intercalated
than thechains,
schematic;
i.e., the cellulose
and the
cellulose/anion
are upon
not completely
straight
and flat,
respectively.
The transformation
the bridging
state
is completed
at is
40 less
mol %,
which than the
by the cations,
forming
a layered
structure. toNote
that the
exact
structure
ordered
enables the IL to thoroughly deconstruct the cellulose crystalline structure at this high concentration.
schematic; i.e., the cellulose chains and the cellulose/anion sheets are not completely straight and flat,
The bridging state is maintained above 40 mol %, where undissolved cellulose co-exists. Note that in
respectively.
transformation
the bridging
statemixtures
is completed
40 mol %,
which enables
the The
concentration
range ofto 10–70
mol %, the
exhibit at
a lyotropic
cholesteric
liquid the IL to
thoroughlycrystallinity,
deconstruct
the cellulose
structure at this high concentration. The bridging state
associated
with the crystalline
chain alignment.
is maintained above 40 mol %, where undissolved cellulose co-exists. Note that in the concentration
Although these findings provided a new aspect of cellulose dissolution state in IL, knowledge is
range of 10–70 mol %, the mixtures exhibit a lyotropic cholesteric liquid crystallinity, associated with
limited at the molecular level. It is well known that cellulose exhibits a hierarchical structure [10–12]:
the chain alignment.
Although these findings provided a new aspect of cellulose dissolution state in IL, knowledge is
limited at the molecular level. It is well known that cellulose exhibits a hierarchical structure [10–12]:
Molecules 2017, 22, 178
Molecules 2017, 22, 178
3 of 9
3 of 9
Molecules
2017, 22, 178
3 ofform
9
nano-level,
forms microfibrils
thethe
diameter
of several
nm, which
gathergather
together
to
at at
thethe
nano-level,
itit forms
microfibrilswith
with
diameter
of several
nm, which
together
to
larger
bundles.
From From
this standpoint,
this study
is devoted
investigating
the structuralthe
changes
in
form
larger
bundles.
this standpoint,
this
study istodevoted
to investigating
structural
at the nano-level, it forms microfibrils with the diameter of several nm, which gather together to form
cellulose/[Emim][OAc]
at the nano-level—particularly
focusing onfocusing
the high on
concentration
regime—
changes
in bundles.
cellulose/[Emim][OAc]
at the
nano-level—particularly
the highchanges
concentration
larger
From this standpoint,
this
study is devoted to investigating
the structural
in
by means means
of smallofangle
X-ray
scattering
(SAXS) measurements,
supportedsupported
by scanning
regime—by
small
angle
X-ray scattering
(SAXS)
measurements,
byelectron
scanning
cellulose/[Emim][OAc]
at the
nano-level—particularly
focusing
on the high concentration
regime—
microscopy.
electron
microscopy.
by means
of small angle X-ray scattering (SAXS) measurements, supported by scanning electron
microscopy.
2. Results
and
Discussion
2. Results
and
Discussion
2. Results
and
Discussion
2 versus
Figure
shows
SAXSpatterns
patternsofofcellulose/[Emim][OAc]
cellulose/[Emim][OAc] in
q2 qversus
q) q)
Figure
2 2shows
SAXS
in the
theKratky
Kratkyplot
plot(Intensity
(Intensity× ×
2
0–100
mol%%
concentration.
In
scattering
from
nanostructures
appears
as a× peak
Figure
2concentration.
shows
SAXS patterns
ofplot,
cellulose/[Emim][OAc]
in nanostructures
the Kratky plot
(Intensity
qasversus
q)For
at at
0–100
mol
Inthis
this
plot,
scattering
from
appears
a[13].
peak
[13].
pure
cellulose
(100
mol
%),
only
a
shoulder
is
observable
in
the
plot,
indicating
the
presence
of of
at
0–100
mol
%
concentration.
In
this
plot,
scattering
from
nanostructures
appears
as
a
peak
[13].
For
For pure cellulose (100 mol %), only a shoulder is observable in the plot, indicating the presence
structures
larger than
the detectable
of theiscurrent
SAXSinsetup
(>50 nm).
shoulder
scattering
pure cellulose
(100 mol
%), only alimit
shoulder
observable
the plot,
the presence
of
structures
larger
than
the
detectable
limit
of
the current
SAXS
setup
(>50indicating
nm).The
The
shoulder
scattering
structures
the detectable
limitmicrofibrils,
of the currentthe
SAXS
setup (>50
nm). The
scattering
would
comelarger
fromthan
bundles
of cellulose
presence
of which
wasshoulder
confirmed
by SEM
would come from bundles of cellulose microfibrils, the presence of which was confirmed by SEM
would come
from bundles
of cellulose
presence
of which
was
confirmed by SEM
observation
(far-right
in Figure
3). Thismicrofibrils,
shoulder isthenot
observed
for the
cellulose/[Emim][OAc]
observation
(far-right
in in
Figure
3).3).This
shoulder
is
not observed
observedfor
forthe
thecellulose/[Emim][OAc]
cellulose/[Emim][OAc]
observation
(far-right
Figure
This
shoulder
is
not
mixtures up to a concentration of 40 mol %, even though the upper limit of dissolution at room
mixtures
up to
a concentration
ofof4040mol
%,
even though
though theupper
upper
limit
of dissolution
at room
mixtures
up
a concentration
molto
%,
limit
of dissolution
at room
temperature
fortothis
system was reported
beeven
26–28 mol %the
[14,15]. This
behavior
is consistent
with
temperature
for this
system
was reported
mol%%[14,15].
[14,15].
This
behavior
is consistent
with
for this
system
reportedto
tobe
be 26–28
26–28
mol
This
behavior
consistent
thetemperature
findings obtained
at thewas
molecular
level
[9],
and
can be explained
by theis concept
ofwith
anion
thebridging.
findings
obtained
at
the
molecular
level
[9],
and
can
be
explained
by
the
concept
of
anion
bridging.
the findings
obtained
at
the
molecular
level
[9],
and
can
be
explained
by
the
concept
of
anion
Namely, the anion bridging form—which is energetically preferred around a concentration
Namely,
the anion
bridging form—which
is energetically
preferred
around
a concentration
Namely,
the anion
bridging
form—which
is energetically
preferred
around
a concentration
of 40
mol
of bridging.
40 mol
%
(and
above)—can
completely
deconstruct
the crystalline
structure
of cellulose,
and
of
40
mol
%
(and
above)—can
completely
deconstruct
the
crystalline
structure
of
cellulose,
and
% (and
above)—can
completely
deconstruct
the
crystalline
structure
of
cellulose,
and
consequently,
consequently, the fiber structures at this concentration. These SAXS results do not contradict the SEM
consequently,
the
structures
at this concentration.
These SAXS
results
do not contradict
the SEM
theobservations
fiber
structures
atfiber
this
concentration.
Theseonly
SAXS
results
do not
contradict
the
observations
shown
in Figure
3, while SEM
detects
surface
morphology.
TheSEM
bundled
fibers
observations shown in Figure 3, while SEM only detects surface morphology. The bundled fibers
shown
Figure at
3, 70–100
while SEM
only
detects
Thethan
bundled
fibers were
observed
were in
observed
mol %.
At 60
mol %,surface
surfacemorphology.
roughness rather
fiber structures
was
seen.
were observed at 70–100 mol %. At 60 mol %, surface roughness rather than fiber structures was seen.
at 70–100
mol %.
At 60 smoother
mol %, surface
roughness cellulose
rather than
fiber structures
seen. The
The surface
became
with decreasing
concentration,
and was
eventually
no surface
clear
The surface became smoother with decreasing cellulose concentration, and eventually no clear
became
smoother
with
decreasing
cellulose
concentration,
and
eventually
no
clear
structure
was
structure
was
detected
at
30
mol
%.
structure was detected at 30 mol %.
detected at 30 mol %.
Figure 2. Small angle X-ray scattering (SAXS) patterns of the cellulose/[Emim][OAc] mixtures in the
Figure
2. 2.
Small
angle
X-ray
mixtures
Figure
Small
angle
X-rayscattering
scattering(SAXS)
(SAXS)patterns
patterns of
of the
the cellulose/[Emim][OAc]
cellulose/[Emim][OAc] mixtures
in in
thethe
Kratky plot.
Kratky
plot.
Kratky plot.
Figure 3. SEM micrographs. White scale bars for the top (100×) and bottom (100,000×) figures
represent
300 μm
and 300 nm,White
respectively.
Figure
SEM
micrographs.
scale bars
bars for
for the
the top
top (100
(100×)
Figure
3. 3.SEM
micrographs.
White scale
×) and
and bottom
bottom(100,000×)
(100,000×figures
) figures
represent
300
μm
and
300
nm,
respectively.
represent 300 µm and 300 nm, respectively.
Molecules 2017, 22, 178
4 of 9
There is another characteristic feature in Figure 1, which is the emergence of two new peaks in
the concentration range of 30–80 mol %. This clearly indicates the presence of two nanostructures (or
aggregations) which are not present in either intact cellulose or [Emim][OAc]. To derive their size
and conduct a detailed discussion, the peaks at 1.1 and 0.3 nm−1 were fitted with the generalized
Guinier–Porod function [16]. The generalized Guinier–Porod function experimentally combines the
Guinier and Porod regions, and is expressed as:
!
−q2 R2g
I (0)
I (q) = s exp
for q ≤ q1 ,
q
3−s
I (q) =
where
D
for q ≥ q1 ,
qd
1
1 (d − s)(3 − s) 2
q1 =
,
Rg
2
2 2
−q R
(d−s)
D = I (0) exp 3−1 s g q1
h
i d−s ,
s) (d−s)(3−s) 2
= I((d0−)s) exp − (d−
2
2
(1)
(2)
(3)
(4)
Rg
and where I(0) is the scattered intensity at q = 0 and Rg is the radius of gyration. A value of s = 0
occurs when the scatterer is a three-dimensional globular object (such as a sphere). For one- (rod) and
two-dimensional (lamellae or platelet) objects, the values are defined as 1 and 2, respectively.
The fitted results are displayed in Figure 4a. Note that the Porod function was employed for the
shoulder scatterer; therefore, the sum of two generalized Guinier–Porod functions and one Porod
function was utilized to reproduce the overall patterns. All experimental patterns at the concentrations
of 30–80 mol % were well reproduced in the entire q range by the combination of the functions. The
derived parameters are summarized in Table 1. The s values for the aggregates were mostly zero,
strongly implying that the shape of these structures was close to a sphere. The diameter of a sphere (R)
can be determined from Rg as
5
R2 = R2g
(5)
3
The concentration dependence of the R values is shown in Figure 4b. The behavior of the two
structures with varying concentration is quite different. The structure with smaller R exists only at
50–80 mol %, and its size is independent of the concentration. In contrast, the structure with larger R is
observed at 30–80 mol %, exhibiting concentration dependency. The invariant 4.1 nm diameter for
the smaller scatterer is reminiscent of the microfibril of cellulose, whose size was reported as 4–5 nm
for Avicel [17,18]. It is well known that cellulose fibrillation occurs with mechanical treatment in the
presence of solvents [19]. The association of the 4.1-nm-diameter scatterer with the cellulose microfibril
is in line with the fact that the shoulder scatterer associated with large bundles also does not exist at
concentrations below 50 mol %. If so, then the result demonstrates that the interfibril breakages occur
more easily than the intrafibril ones. The dynamic dissolution process can proceed similarly.
Molecules 2017, 22, 178
Molecules 2017, 22, 178
5 of 9
5 of 9
Figure 4. (a) SAXS patterns of the cellulose/[Emim][OAc] mixtures in the Kratky plot for the
Figure 4. (a) SAXS patterns of the cellulose/[Emim][OAc] mixtures in the Kratky plot for the
concentrations of 30–80 mol %. Black curves are the observed patterns, while the red curves are the
concentrations of 30–80 mol %. Black curves are the observed patterns, while the red curves are
theoretical patterns comprising the sum of two generalized Guinier–Porod functions and one Porod
the theoretical
patterns comprising the sum of two generalized Guinier–Porod functions and one Porod
function; (b) Red: Diameter changes of the aggregated structures, assuming a sphere. Filled and open
function;
(b)
Red:
Diameter
changes
of the aggregated
structures,
assuming
a sphere.
and open
symbols are for
large and
small aggregates,
respectively.
Black: The
normalized
I2(0)/I1Filled
(0) change,
symbols
are
for
large
and
small
aggregates,
respectively.
Black:
The
normalized
I
(0)/I
(0)
change,
2 1 and
1 2 are
where I1(0) and I2(0) are the scattered intensities of the aggregates at q = 0. The subscripts
whereforI1the
(0) and
I
(0)
are
the
scattered
intensities
of
the
aggregates
at
q
=
0.
The
subscripts
1
and
large 2and small aggregates, respectively. The ± signs represent standard deviation produced 2 are
in large
the fitting
based on
a nonlinear regression
method.
for the
and procedures
small aggregates,
respectively.
The ± signs
represent standard deviation produced in
the fitting procedures based on a nonlinear regression method.
Table 1. SAXS parameters obtained from Figure 4a. The subscripts 1, 2, and s are for the large
aggregates, small aggregates, and shoulder scattering, respectively. The ± signs represent standard
Table 1. SAXS parameters obtained from Figure 4a. The subscripts 1, 2, and s are for the large
deviation produced in the fitting procedures based on a nonlinear regression method.
aggregates, small aggregates, and shoulder scattering, respectively. The ± signs represent standard
deviationCellulose
produced in the Rfitting
procedures
based
on a nonlinear
regression
method.
g1/nm
d1
s1
Rg2/nm
d2
s2
ds
Concentration/mol %
7.9 ± 0.4
Cellulose30
Rg1 /nm
35 %
7.5 ± 0.2
Concentration/mol
40
7.1 ± 0.1
30
7.9 ± 0.4
50
35
7.5 ±5.5
0.2± 1.5
60
40
7.1 ±5.9
0.1± 0.5
70
50
5.5 ±4.9
1.5± 0.2
60
5.9 ±4.7
0.5± 0.0
80
70
4.9 ± 0.2
80
4.7 ± 0.0
4.0 ± 0.0
d1 ± 0.0
3.8
3.9 ± 0.0
4.0 ± 0.0
0.3
3.83.0
± ±0.0
0.0
3.94.0
± ±0.0
0.8
3.03.6
± ±0.3
4.03.8
± ±0.0
0.4
3.6 ± 0.8
3.8 ± 0.4
0.0 ± 0.1
0.0 s±1 0.1
0.0 ± 0.1
0.0 ± 0.1
0.0 ±
± 0.8
0.0
0.1
0.0 ±
± 0.0
0.0
0.1
0.1 ±
± 0.1
0.0
0.8
0.0
0.0
0.1 ±
± 0.1
0.1 ± 0.1
0.1 ± 0.1
n/a
Rn/a
g2 /nm
n/a
n/a
1.6 n/a
± 0.5
1.6 n/a
± 0.0
1.6
1.6±±0.0
0.5
1.6±±0.1
0.0
1.5
1.6 ± 0.0
1.5 ± 0.1
n/a
n/ad2
n/a
n/a
3.5 ±n/a
0.4
4.0 ±n/a
0.0
4.03.5
± 0.0
± 0.4
± 0.0
4.04.0
± 0.0
4.0 ± 0.0
4.0 ± 0.0
n/a
n/a
n/a s2
n/a ds
n/a
n/a
n/a
n/a
0.0 ± 0.9
4.0
± 1.5n/a
n/a
0.1 ± 0.1
n/a 4.0 ± 0.0n/a
0.1 0.0
± 0.1
0.0 ± 1.5
± 0.9 4.0 ± 4.0
± 0.1 4.0 ± 4.0
0.1 0.1
± 0.1
0.0 ± 0.0
0.1 ± 0.1
4.0 ± 0.0
0.1 ± 0.1
4.0 ± 0.0
The next question is on the origin of the larger scatterer. During the dissolution process, MD
simulations observe the peeling of cellulose chains on the fibril surface from the chain edges by
intercalation
of IL ions
Since of
thethe
intramolecular
hydrogen
bonding
of peeled chains
is MD
The
next question
is on[20,21].
the origin
larger scatterer.
During
the dissolution
process,
considerably
disrupted
by
the
IL
ions,
chain
bending
and
twisting
occur
[22,23].
This
flexibility
of
the
simulations observe the peeling of cellulose chains on the fibril surface from the chain edges by
terminal parts
such Since
chains the
to entangle,
which could
correspond
to the larger
aggregate.
intercalation
of ILmay
ionsenable
[20,21].
intramolecular
hydrogen
bonding
of peeled
chains is
This structure disappears at 25 mol %, because all OH groups can interact with an anion in a 1:1 ratio
considerably disrupted by the IL ions, chain bending and twisting occur [22,23]. This flexibility
(non-bridging) at this concentration, which unravels the entanglement. Another piece of indirect
of the terminal parts may enable such chains to entangle, which could correspond to the larger
evidence for the assignment of this structure is the moldability of the cellulose/IL mixture. After
aggregate.
structure
disappears
at 25 mol %, because
all OH
groups
can%interact
withbecame
an anion in
severalThis
hours
of mixing
cellulose/[Emim][OAc],
the samples
of 30–50
mol
intriguingly
a 1:1 mechanically
ratio (non-bridging)
at this
unravels
Another piece
sufficiently
hardconcentration,
to be moldedwhich
(Figure
5), and the
thisentanglement.
moldability disappeared
at of
indirect
evidence for
the assignment
of idea
this that
structure
is the
moldability
cellulose/IL
mixture.
concentrations
below
30 mol %. The
cellulose
entanglement
is of
thethe
origin
of mechanical
well-established
cellulose hydrogels [24,25].the
Although
exploring
themol
potential
of this
Afterstrength
severalishours
of mixingfor
cellulose/[Emim][OAc],
samples
of 30–50
% intriguingly
cellulose/IL
composite
film is beyond
of this paper,
it is
that among reported
became
mechanically
sufficiently
hardthe
to scope
be molded
(Figure
5),worth
and noting
this moldability
disappeared
cellulose/IL composite
materials
the that
emergence
of the
moldability in is
cellulose/IL
with
at concentrations
below 30
mol %.[26–29],
The idea
cellulose
entanglement
the originsystem
of mechanical
liquid
crystallinity
[9]
(in
a
10–70
mol
%
regime)
has
been
discovered
here
for
the
first
time.
The
strength is well-established for cellulose hydrogels [24,25]. Although exploring the potential
of this
aggregate size decreases with increasing cellulose concentration (Figure 4b), probably because of the
cellulose/IL composite film is beyond the scope of this paper, it is worth noting that among reported
decrease in the numbers of glucose units in the chains and the IL ions that participate in the aggregates.
cellulose/IL composite materials [26–29], the emergence of the moldability in cellulose/IL system
with liquid crystallinity [9] (in a 10–70 mol % regime) has been discovered here for the first time. The
aggregate size decreases with increasing cellulose concentration (Figure 4b), probably because of the
decrease in the numbers of glucose units in the chains and the IL ions that participate in the aggregates.
Molecules 2017, 22, 178
6 of 9
Molecules 2017, 22, 178
6 of 9
Molecules 2017, 22, 178
6 of 9
Figure
5. Cellulose/IL
(ionicliquid)
liquid)composite
composite films
%)%)
with
a diameter
of 25ofmm
and and
Figure
5. Cellulose/IL
(ionic
films(30–50
(30–50mol
mol
with
a diameter
25 mm
thickness
of 1 mm.
opacity
thefilm
filmofof40
40 mol
mol %
micro
airair
bubbles
[9] and
surface
thickness
of 1 mm.
TheThe
opacity
of of
the
% isisattributed
attributedtoto
micro
bubbles
[9] and
surface
(or interfacial)
roughness, which
were confirmed
infilms
the 30(30–50
and
theaSEM
images
3and
Figure roughness,
5. Cellulose/IL
(ionicwere
liquid)
composite
%)
with
of(Figure
25 mm
(or interfacial)
which
confirmed
in
the 30mol
mol%mol
%film
film
and
thediameter
SEM
images
(Figure
3
top).thickness
Conversely,
the
presence
of
undissolved
crystals
is
also
a
reason
for
the
opacity
at
50
mol
%.
of
1
mm.
The
opacity
of
the
film
of
40
mol
%
is
attributed
to
micro
air
bubbles
[9]
and
surface
top). Conversely, the presence of undissolved crystals is also a reason for the opacity at 50 mol %.
(or interfacial) roughness, which were confirmed in the 30 mol % film and the SEM images (Figure 3
Thetop).
mixture
with 25
% cellulose—which
gives is
some
no peak
Conversely,
themol
presence
of undissolved crystals
also ascattering
reason for but
the opacity
at in
50 the
mol Kratky
%.
The
mixture be
with
25 mol
gives some
scattering
butwould
no peak
the Kratky
plot—cannot
fitted
with %
thecellulose—which
generalized Guinier–Porod
model,
and thus
needinanother
theoretical
etthe
al.mol
[15]
the
SAXS gives
profilesome
of
the
transparent
gel-like
of
plot—cannot
bemodel.
fittedRein
with
generalized
Guinier–Porod
model,
and thus
would
need
another
The
mixture
with
25
%reported
cellulose—which
scattering
but no
peak mixture
in
the Kratky
cellulose/[Emim][OAc]
at
28
mol
%,
which
was
prepared
by
rapid
evaporation
of
a
volatile
plot—cannot
be
fitted
with
the
generalized
Guinier–Porod
model,
and
thus
would
need
another
theoretical model. Rein et al. [15] reported the SAXS profile of the transparent gel-like mixture of
co-solvent.
They
applied
aet combination
of prepared
the
[30]
and Ornstein–Zernike
[31] of
theoretical
model.
al.
reported
the Debye–Bueche
SAXS by
profile
the transparent
cellulose/[Emim][OAc]
atRein
28 mol
%,[15]
which
was
rapidofevaporation
of agel-like
volatilemixture
co-solvent.
equations
for
the
pattern
to
gain
correlation
lengths
of
the
scatterer.
The
former
and
latter
equations
cellulose/[Emim][OAc]
at the
28 Debye–Bueche
mol %, which was
by rapid evaporation
of a volatile
They applied
a combination of
[30] prepared
and Ornstein–Zernike
[31] equations
for the
express
the scattering
from a long-range
inhomogeneity
and a fluctuation
in polymer
chains, and[31]
co-solvent.
They applied
a combination
of the Debye–Bueche
[30] and
Ornstein–Zernike
pattern to gain correlation lengths of the scatterer. The former and latter equations express the scattering
were
employed
to describe
major
and minor
components
in theThe
pattern,
The
equations
for the
pattern tothe
gain
correlation
lengths
of the scatterer.
formerrespectively.
and latter equations
from Debye–Bueche
a long-range inhomogeneity
and a fluctuation in polymer chains, and were employed to describe
equation
is
express the scattering from a long-range inhomogeneity and a fluctuation in polymer chains, and
the major
andemployed
minor components
themajor
pattern,
The Debye–Bueche
is The
were
to describeinthe
and respectively.
minor
components
in the pattern,equation
respectively.
A
(
)
I
q
=
2
Debye–Bueche equation is
(6)
1 + ξ 2 q 2A ,
,
(6)
I (q) =
2
A
I (q ) =(1 + ξ 2 q22)
(6)
where ξ is the correlation length. The Ornstein–Zernike
is
1 + ξ 2 q 2equation
,
(
)
(
)
where ξ is the correlation length. The Ornstein–Zernike
equation is
(q) = A2 2
where ξ is the correlation length. TheIOrnstein–Zernike
equation is
(7)
1+ ξ q .
A
A
I (qI)(q=
.
(7)
) =1require
2 2
The quality of our data for 25 mol % does
not
+
1 + ξξ2 qq2 .the Ornstein–Zernike equation, and yields(7)
a correlation length of 4.8 nm with the Debye–Bueche equation (Figure 6). This value is smaller than
The
quality
of
ourof
data
for
25for
mol
does
not require
thethe
Ornstein–Zernike
equation,
yields
the reported
value
nm
[15],
possibly
owing
to
slightly
lower
cellulose concentration
of our
The quality
of18
our
data
25%
mol
%
does
notthe
require
Ornstein–Zernike
equation,and
and
yields a
correlation
length
4.8 nm
with
the
Debye–Bueche
equation
(Figure
6).6).This
issmaller
smaller
than
sample
and theoflength
differences
in
sample
preparation
procedures,
whereas
the value
differences
in the
a correlation
of 4.8
nmthe
with
the Debye–Bueche
equation
(Figure
This
value is
than
fitting
of of
the
SAXS
pattern
andpossibly
dataowing
quality
S/N
ratio
for our
results)concentration
could contribute
the reported
value
18
nm
[15],
to(lower
the
slightly
lower
cellulose
our
the range
reported
value
of 18
nm possibly
[15],
owing
to the
slightly
lower
cellulose
concentration
ofof
our
to the
gap
thethe
correlation
length.
In any
case, preparation
it is indicated
that the
25 mol
% sample
does not
form
sample
differences
in the
sample
procedures,
whereas
the differences
infitting
the
sample
and
theofand
differences
in the
sample
preparation
procedures,
whereas
the
differences
in the
structures,
which
seen
in
the
samples
with
higher
cellulose
range
of the
SAXS
pattern
and
data
quality
(lower
S/N
ratio
forconcentrations.
our
results)
could contribute
rangedefinite
offitting
the SAXS
pattern
andwere
data
quality
(lower
S/N
ratio
for
our
results)
could
contribute
to the gap
to
the
gap
of
the
correlation
length.
In
any
case,
it
is
indicated
that
the
25
mol
%
sample
notdefinite
form
of the correlation length. In any case, it is indicated that the 25 mol % sample does notdoes
form
definite structures, which were seen in the samples with higher cellulose concentrations.
structures, which were seen in the samples with higher cellulose concentrations.
Figure 6. A single logarithmic SAXS pattern of the mixture at 25 mol %. The Debye–Bueche equation
was applied (red line) to reproduce the experimental pattern.
6. Alogarithmic
single logarithmic
pattern
of the
mixture
at 25
mol
TheDebye–Bueche
Debye–Bueche equation
Figure 6.Figure
A single
SAXSSAXS
pattern
of the
mixture
at 25
mol
%.%.The
equation
was applied (red line) to reproduce the experimental pattern.
was applied (red line) to reproduce the experimental pattern.
Molecules 2017, 22, 178
7 of 9
3. Materials and Methods
Microcrystalline cellulose (Avicel PH-101, Sigma-Aldrich, St. Louis, MO, USA) and [Emim][OAc]
(>95%, IoLiTec, Heilbronn, Germany) were dried by heating under vacuum (1 Pa) before use. Avicel
PH-101 contains approximately 5 wt % of water, and after drying at 343 K for 3 h, the level of water
detectable by a moisture analyzer (Sartorius, model MA35) was zero. The water content in the IL was
measured using 1 H-NMR (JEOL JNM-ECS400 or JNM-ECA600, Tokyo, Japan), and no water peak
signal was detected after one day of drying at 323 K. The typical sample-mixing procedure was as
follows: 0.1 g of Avicel was added to a certain amount of the IL in a vial and mixed with a spatula
until the mixture appeared visually homogenous (normally after 5 min). The mixtures were prepared
in the range of 10–90 mol % at every 5 or 10 mol %. After one day (which is considered a sufficient
period to reach an experimental equilibrium state [9]), the mixture was packed within a Teflon cell
with 1 mm thickness. The cell was sandwiched between two sheets of Kapton film (5 µm thickness)
and sealed with grease (Apiezon N). These procedures were performed in an inert-atmosphere glove
box (the water content was less than 10 ppm). In the mixtures prepared with the current condition, no
detectable chemical reactions or cellulose degradation occurred [9].
SAXS measurements were conducted with a NANO-Viewer (IP system, Rigaku Co., Ltd,
Tokyo, Japan) with a Cu Kα radiation source (λ = 0.154 nm) at an applied voltage of 40 kV and
a filament current of 30 mA. The distance between the sample and the detector was set at 700 mm
(q = 0.142–2.844 nm−1 ). The samples were irradiated for 1 h. The obtained two-dimensional image was
transformed into a one-dimensional pattern for theoretical analyses. It should be noted that anisotropic
images were not observed in any of our data because the cellulose sample used here was powder. The
scattering experiments were conducted three times with a cellulose concentration of more than 40 mol
% at different spots in the same sample, and the averaged patterns were used for analyses. This is
because there exists inhomogeneity in the sample with such high concentrations due to the presence of
undissolved (intact cellulose) regions [9]. Backgrounds including scattering from Kapton films were
subtracted from the scattering patterns.
A field emission scanning electron microscope (FE-SEM, JEOL JSM-7100F) was operated at 15 kV.
The samples were coated with Au/Pt using an ion sputter coater for SEM observation.
4. Conclusions
We have investigated structural changes of cellulose/[Emim][OAc] at the nano-level over the
entire concentration range of 0–100 mol % by SAXS measurements with the aid of SEM observations.
The shoulder scattering in the Kratky plot ascribed to bundles of cellulose microfibrils does not exist
up to 40 mol %. This finding is well explained by the fact that at high concentrations, the anions
are energetically preferred to have multiple interactions with OH groups of cellulose chains (anion
bridging), which enables the deconstruction of the cellulose crystallinity, and consequently fibril
structures at 40 mol %. We observed two aggregated structures that are not observed either in intact
cellulose or in the IL at 30–80 mol %. The smaller structure—which is reminiscent of a microfibril with
an average diameter of 4.1 nm—does not show concentration dependency. Conversely, the larger one
depends on the concentration, and its size is in the range of 12–20 nm. The structure originates from
cellulose chain entanglement, which is the origin of the moldability of the cellulose/IL composite films
that possess liquid crystallinity.
Since the local concentration of cellulose in ILs at the early stage of the dissolution process
should be much higher than that of the consequent solution, the nano-structures observed in the high
concentrations can be intermediate states for the dissolution process. Therefore, our findings contribute
to an understanding of the dissolution mechanism of cellulose in the IL; inter-fibrillation occurs more
easily than intra-fibrillation, and cellulose chains are entangled with each other after IL ions peel the
chains from fibrils, which may be one reason for the slow dissolution process of cellulose in ILs.
Molecules 2017, 22, 178
8 of 9
Acknowledgments: This study was supported in part by Advanced Low Carbon Technology Research and
Development Program (ALCA) (Grant number 2100040), Cross-ministerial Strategic Innovation Promotion
Program (SIP), and Center of Innovation Program “Construction of next-generation infrastructure using innovative
materials: Realization of safe and secure society that can coexist with the Earth for centuries” from Japan Science
and Technology Agency, and also by JSPS KAKENHI Grant Number 15H04193.
Author Contributions: T.E. and K.T. conceived and designed the experiments; S.H. and S.F. performed the
experiments; T.E. analyzed the data; T.E. wrote the paper in consultation with K.T. and K.N.
Conflicts of Interest: The authors declare no conflict of interest.
References
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
Swatloski, R.P.; Spear, S.K.; Holbrey, J.D.; Rogers, R.D. Dissolution of Cellose with Ionic Liquids. J. Am.
Chem. Soc. 2002, 124, 4974–4975. [CrossRef] [PubMed]
Pinkert, A.; Marsh, K.N.; Pang, S.; Staiger, M.P. Ionic liquids and their interaction with cellulose. Chem. Rev.
2009, 109, 6712–6728. [CrossRef] [PubMed]
Wang, H.; Gurau, G.; Rogers, R.D. Ionic liquid processing of cellulose. Chem. Soc. Rev. 2012, 41, 1519–1537.
[CrossRef]
Gupta, K.M.; Jiang, J. Cellulose dissolution and regeneration in ionic liquids: A computational perspective.
Chem. Eng. Sci. 2015, 121, 180–189. [CrossRef]
Yuan, X.; Cheng, G. From cellulose fibrils to single chains: Understanding cellulose dissolution in ionic
liquids. Phys. Chem. Chem. Phys. 2015, 17, 31592–31607. [CrossRef] [PubMed]
Remsing, R.C.; Swatloski, R.P.; Rogers, R.D.; Moyna, G. Mechanism of cellulose dissolution in the ionic
liquid 1-n-butyl-3-methylimidazolium chloride: A 13 C and 35/37 Cl NMR relaxation study on model systems.
Chem. Commun. 2006, 1271, 1271–1273. [CrossRef] [PubMed]
Youngs, T.G.A.; Holbrey, J.D.; Mullan, C.L.; Norman, S.E.; Lagunas, M.C.; D’Agostino, C.; Mantle, M.D.;
Gladden, L.F.; Bowron, D.T.; Hardacre, C. Neutron diffraction, NMR and molecular dynamics study of
glucose dissolved in the ionic liquid 1-ethyl-3-methylimidazolium acetate. Chem. Sci. 2011, 2, 1594–1605.
[CrossRef]
Rabideau, B.D.; Agarwal, A.; Ismail, A.E. The role of the cation in the solvation of cellulose by
imidazolium-based ionic liquids. J. Phys. Chem. B 2014, 118, 1621–1629. [CrossRef] [PubMed]
Endo, T.; Hosomi, S.; Fujii, S.; Ninomiya, K.; Takahashi, K. Anion Bridging-Induced Structural Transformation
of Cellulose Dissolved in Ionic Liquid. J. Phys. Chem. Lett. 2016, 7, 5156–5161. [CrossRef] [PubMed]
Cheng, G.; Zhang, X.; Simmons, B.; Singh, S. Theory, practice and prospects of X-ray and neutron
scattering for lignocellulosic biomass characterization: Towards understanding biomass pretreatment.
Energy Environ. Sci. 2015, 8, 436–455. [CrossRef]
Nishiyama, Y. Structure and properties of the cellulose microfibril. J. Wood Sci. 2009, 55, 241–249. [CrossRef]
Martínez-Sanz, M.; Gidley, M.J.; Gilbert, E.P. Application of X-ray and neutron small angle scattering
techniques to study the hierarchical structure of plant cell walls: A review. Carbohydr. Polym. 2015, 125,
120–134. [CrossRef] [PubMed]
Doniach, S. Changes in biomolecular conformation seen by small angle X-ray scattering. Chem. Rev. 2001,
101, 1763–1778. [CrossRef] [PubMed]
Le, K.A.; Rudaz, C.; Budtova, T. Phase diagram, solubility limit and hydrodynamic properties of cellulose in
binary solvents with ionic liquid. Carbohydr. Polym. 2014, 105, 237–243. [CrossRef] [PubMed]
Rein, D.M.; Khalfin, R.; Szekely, N.; Cohen, Y. True molecular solutions of natural cellulose in the binary
ionic liquid-containing solvent mixtures. Carbohydr. Polym. 2014, 112, 125–133. [CrossRef] [PubMed]
Hammouda, B. A new Guinier-Porod model. J. Appl. Crystallogr. 2010, 43, 716–719. [CrossRef]
Elazzouzi-Hafraoui, S.; Nishiyama, Y.; Putaux, J.L.; Heux, L.; Dubreuil, F.; Rochas, C. The shape and size
distribution of crystalline nanoparticles prepared by acid hydrolysis of native cellulose. Biomacromolecules
2008, 9, 57–65. [CrossRef] [PubMed]
Leppänen, K.; Andersson, S.; Torkkeli, M.; Knaapila, M.; Kotelnikova, N.; Serimaa, R. Structure of cellulose
and microcrystalline cellulose from various wood species, cotton and flax studied by X-ray scattering.
Cellulose 2009, 16, 999–1015. [CrossRef]
Isogai, A. Wood nanocelluloses: Fundamentals and applications as new bio-based nanomaterials. J. Wood Sci.
2013, 59, 449–459. [CrossRef]
Molecules 2017, 22, 178
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
9 of 9
Rabideau, B.D.; Agarwal, A.; Ismail, A.E. Observed Mechanism for the Breakup of Small Bundles of Cellulose
Iα and Iβ in Ionic Liquids from Molecular Dynamics Simulations. J. Phys. Chem. B 2013, 117, 3469–3479.
[CrossRef] [PubMed]
Cho, H.M.; Gross, A.S.; Chu, J.W. Dissecting force interactions in cellulose deconstruction reveals the required
solvent versatility for overcoming biomass recalcitrance. J. Am. Chem. Soc. 2011, 133, 14033–14041. [CrossRef]
[PubMed]
Liu, H.; Sale, K.L.; Holmes, B.M.; Simmons, B.A.; Singh, S. Understanding the interactions of cellulose with
ionic liquids: A molecular dynamics study. J. Phys. Chem. B 2010, 114, 4293–4301. [CrossRef] [PubMed]
Li, Y.; Liu, X.; Zhang, S.; Yao, Y.; Yao, X.; Xu, J.; Lu, X. Dissolving process of a cellulose bunch in ionic liquids:
A molecular dynamics study. Phys. Chem. Chem. Phys. 2015, 17, 17894–17905. [CrossRef] [PubMed]
Ando, H.; Konishi, T. Structure analysis of regenerated cellulose hydrogels by small-angle and
ultra-small-angle X-ray scattering. Phys. Rev. E 2000, 62, 727–733. [CrossRef]
Ishii, D.; Tatsumi, D.; Matsumoto, T.; Murata, K.; Hayashi, H.; Yoshitani, H. Investigation of the structure of
cellulose in LiCl/DMAc solution and its gelation behavior by small-angle X-ray scattering measurements.
Macromol. Biosci. 2006, 6, 293–300. [CrossRef] [PubMed]
Isik, M.; Sardon, H.; Mecerreyes, D. Ionic liquids and cellulose: Dissolution, chemical modification and
preparation of new cellulosic materials. Int. J. Mol. Sci. 2014, 15, 11922–11940. [CrossRef] [PubMed]
Zhu, S.; Wu, Y.; Chen, Q.; Yu, Z.; Wang, C.; Jin, S.; Ding, Y.; Wu, G. Dissolution of cellulose with ionic liquids
and its application: A mini-review. Green Chem. 2006, 8, 325–327. [CrossRef]
Takada, A.; Kadokawa, J.-I. Fabrication and Characterization of Polysaccharide Ion Gels with Ionic Liquids
and Their Further Conversion into Value-Added Sustainable Materials. Biomolecules 2015, 5, 244–262.
[CrossRef] [PubMed]
Kunchornsup, W.; Sirivat, A. Effects of crosslinking ratio and aging time on properties of physical and
chemical cellulose gels via 1-butyl-3-methylimidazolium chloride solvent. J. Sol-Gel Sci. Technol. 2010, 56,
19–26. [CrossRef]
Debye, P.; Bueche, A.M. Scattering by an inhomogeneous solid. J. Appl. Phys. 1949, 20, 518–525. [CrossRef]
Ornstein, L.S.; Zernike, F. Accidental deviations of density and opalescence at the critical point of a single
substance. Proc. Acad. Sci. Amst. 1914, 17, 793–806.
Sample Availability: Samples of the compounds are commercially available from the company mentioned in
Section 3 Materials and Methods.
© 2017 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).