Vishnu Arumughan1,2, Tiina Nypelö1,3, Merima Hasani1,2,3, Anette Larsson1,2,3,4. 1. Department of Chemistry and Chemical Engineering, Chalmers University of Technology, SE-412 96 Gothenburg, Sweden. 2. AvanCell, Chalmers University of Technology, SE-412 96 Gothenburg, Sweden. 3. Wallenberg Wood Science Center, Chalmers University of Technology, SE-412 96 Gothenburg, Sweden. 4. FibRe─Centre for Lignocellulose-based Thermoplastics, Department of Chemistry and Chemical Engineering, Chalmers University of Technology, SE-412 96 Gothenburg, Sweden.
Abstract
The adsorption of carboxymethylcellulose (CMC) on cellulose surfaces is one of the most studied examples of the adsorption of an anionic polyelectrolyte on a like-charged surface. It has been suggested that divalent ions can act as a bridge between CMC chains and the surface of cellulose and enhance the CMC adsorption: they can, however, also alter the structure of CMCs in the solution. In previous investigations, the influence of cations on solution properties has been largely overlooked. This study investigates the effect of Ca2+ ions on the properties of CMC solutions as well as the influence on cellulose nanofibers (CNFs), which was studied by dynamic light scattering and correlated with the adsorption of CMC on a cellulose surface probed using QCM-D. The presence of Ca2+ facilitated the multichain association of CMC chains and increased the hydrodynamic diameter. This suggests that the adsorption of CMCs at high concentrations of CaCl2 is governed mainly by changes in solution properties rather than by changes in the cellulose surface. Furthermore, an entropy-driven mechanism has been suggested for the adsorption of CMC on cellulose. By comparing the adsorption of CMC from H2O and D2O, it was found that the release of water from the cellulose surface is driving the adsorption of CMC.
The adsorption of carboxymethylcellulose (CMC) on cellulose surfaces is one of the most studied examples of the adsorption of an anionic polyelectrolyte on a like-charged surface. It has been suggested that divalent ions can act as a bridge between CMC chains and the surface of cellulose and enhance the CMC adsorption: they can, however, also alter the structure of CMCs in the solution. In previous investigations, the influence of cations on solution properties has been largely overlooked. This study investigates the effect of Ca2+ ions on the properties of CMC solutions as well as the influence on cellulose nanofibers (CNFs), which was studied by dynamic light scattering and correlated with the adsorption of CMC on a cellulose surface probed using QCM-D. The presence of Ca2+ facilitated the multichain association of CMC chains and increased the hydrodynamic diameter. This suggests that the adsorption of CMCs at high concentrations of CaCl2 is governed mainly by changes in solution properties rather than by changes in the cellulose surface. Furthermore, an entropy-driven mechanism has been suggested for the adsorption of CMC on cellulose. By comparing the adsorption of CMC from H2O and D2O, it was found that the release of water from the cellulose surface is driving the adsorption of CMC.
The
adsorption of polymers on cellulose surfaces has been of interest
for decades due to its applications in papermaking. The rapid development
of layer-by-layer surface modification techniques on cellulose materials
vitalized further the need for understanding polymer adsorption on
cellulose.[1,2] Cationic polyelectrolytes readily adsorb
on negatively charged cellulose surfaces, with interactions governing
entropy gain due to the release of counterions.[3] It has been shown that anionic polyelectrolytes such as
carboxymethylcellulose can also adsorb on cellulose surfaces irrespective
of their expected electrostatic repulsion.[4−8] Carboxymethylcellulose (CMC) is the most versatile
semisynthetic polyelectrolyte derived from cellulose[9,10] and has been used in a wide range of applications, including mineral
processing,[11−13] pharmaceutics,[14] and as
a viscosity modifier.[15] In the paper industry,
the adsorption of CMC on cellulose has been used to improve the wet
mechanical properties of cellulose.[4,5,16] Apart from conventional applications, the adsorption
of CMC on cellulose has been used to create biofunctional interfaces
for diagnostic platforms and bone-healing scaffolds.[6,17,18]Seminal work done by Laine
et al. suggested a cocrystallization
mechanism due to the similar backbone structure of CMC and cellulose.[4] However, the mechanism of adsorption and the
nature of the driving force behind the interaction of CMC with cellulose
surface are still ambiguous. The adsorption of CMC on cellulose surfaces
has been probed using surface-sensitive techniques, such as QCM-D
and SPR, to distinguish the controlling factors of the adsorption
of CMCs on cellulose fibers.[7,8] Kargl et al. studied
the adsorption of CMC on surfaces of varying cellulose character and
suggested the involvement of a specific interaction between the CMC
chains and the cellulose surface:[7] interactions
between them can be modulated by changing the concentration of the
electrolyte in the system.[4,8] The valency of cations
also plays a critical role in the adsorption process: it has been
shown that the presence of Ca2+ ions enhanced CMC adsorption
and that the adsorbed layer was stable compared to that formed in
the presence of Na+ ions. Taking these observations into
consideration, it has been suggested that Ca2+ ions act
as a bridge between a negatively charged CMC and a cellulose surface.[8]Adsorption is a multifaceted interfacial
process where characteristics
of both the polymer solution and the adsorbing surface are important.
Recent computational studies suggest that the adsorption of anionic
polyelectrolytes on like-charged surfaces is driven by multivalent
ion-induced charge inversion, either on adsorbing surface or in polyelectrolytes.[19,20] This has been observed experimentally by Tiraferri et al. for the
case of adsorption of negatively charged polystyrene sulfonate on
a silica substrate.[21] A recent investigation
by Jansson et al. revealed multivalent ion-induced charge inversion
in clay nanoplatelets.[22] However, such
multivalent ion-induced charge inversion has not yet been reported
in the case of cellulose-based materials. The multivalent ions were
shown to screen the surface charge to a large extent and induce interactions
between carboxylated cellulose nanoparticles through a ion–ion
correlation and dispersion interactions.[23] Recent studies on dilute CMCs solutions have shown that the interaction
of polyelectrolytes with multivalent ions can affect the conformation
of polymer chains in solution, supramolecular association, solubility,
and flow behavior.[24,25] The addition of salt typically
reduces the viscosity of polyelectrolyte solutions due to the reduced
expansion of polymer chains in the solution. This has been observed
in CMC in the presence of NaCl. On the other hand, divalent ions increased
the viscosity of CMC solutions significantly; it has been suggested
that multivalent ions cross-link the CMC chain via electrostatic bridging
and that this cross-linked network contributes to the viscosity.[25] Although Sharratt et al. observed the formation
of 20–40 nm clusters of CMC chains in the presence of multivalent
ions,[24] the influence of these structural
changes in solutions on interfacial processes such as adsorption has
not yet been studied systematically.The focus of the present
investigation is to understand the adsorption
process in the presence of Ca2+ ions by considering structural
changes that occur in CMC solutions as well as changes in the cellulose
surface in terms of charge. Furthermore, the adsorption of CMC from
deuterated water and water is compared to test the emerging hypothesis
that the adsorption of nonionic and anionic polymeric systems is driven
by release of structured water from the cellulose surface.[26−28] Considering the ongoing interest in integrating CMC adsorption with
the current pulping process to modify fibers within the production
line,[29] knowledge of the multivalent ion-induced
effect in CMC adsorption would be useful in designing the unit operations
of the industrial process. Moreover, the outcome of this research
contributes to the fundamental understanding of the adsorption of
anionic polyelectrolytes on like-charged surfaces.
Experimental Section
Materials
The cellulose nanofibers
(CNF) with an average diameter of 5 nm and carboxylic content of 31.4
μmol/g from softwood Kraft fibers were obtained from Stora Enso,
Stockholm. The nanocellulose fibers have a residual hemicellulose
content of 14.7% (xylose 8%, arabinose 0.62%, galactose 0.25%, and
mannose 6.1%) and lignin content of 1.1% (Klason lignin 0.35% and
acid-soluble lignin 0.75%). Calcium chloride (CaCl2), deuterium
oxide (D2O), and polyethyleneimine-branched polymers (with
an average Mw 25 kDa) were purchased from Sigma-Aldrich. Carboxymethylcellulose
(CMC), more precisely Blanose 7LPEP with a molecular weight of 90.5
kDa and a degree of substitution (DS) of 0.7 (according to the supplier),
was kindly provided by Ashland.
Methods
Preparation and Characterization of an Ultrathin
Cellulose Model Film for Adsorption Studies
A cellulose model
film was prepared according to the protocol described in our previous
report.[30] A CNF suspension of concentration
1.6 g/L was sonicated and centrifuged; the supernatant-containing
CNFs were spin-coated on SiO2-coated QCM-D sensors supplied
by Q-Sense AB (Gothenburg, Sweden). Prior to the spin coating with
CNF, the sensors were preabsorbed with an anchoring layer of polyethyleneimine
to promote the CNF attachment and film stability. The spin-coated
QCM sensors were dried in an oven at 80 °C for 10 min and stored
in a desiccator.The film morphology and uniformity were analyzed
using atomic force microscopy (INTEGRA Prima setup NT-MDT Spectrum
Instruments, Moscow, Russia). The height profiles of three random
spots on the film were recorded in semicontact mode, and the root-mean-square
roughness was calculated using Gwyddion software to assess the quality
of the film.The water content of the CNF model film was calculated
according
to the procedure reported by Kittle et al.[30] using a QCM-D instrument (Biolin Scientific, Gothenburg, Sweden).
In this procedure, the QCM-D sensors coated with CNFs were placed
in the flow cell, and deionized water was injected into the flow cell
at the rate of 100 μL per minute for 3 h to get a stable baseline.
Then, the solvent was switched into D2O, and an immediate
drop in frequency was observed due to the relatively high density
of D2O. After 10 min, the solvent was again switched to
water, and the shift in the frequency was recorded. The difference
in the frequency changes for the bare sensor, and the CNF-coated sensor
during H2O–D2O exchange was used for
the calculation of water associated with the spin-coated CNF film.
Hydrodynamic Size of CMCs Determined by
Dynamic Light Scattering
The hydrodynamic size of CMCs in
different electrolyte concentrations was determined using DLS (Zetasizer
Nano Zs, Malvern Instruments, U.K.). CMC solutions with a concentration
of 0.2% (w/v) were prepared by dissolving CMC in Milli-Q water and
stirring overnight to ensure complete dissolution. The electrolyte
concentration in the solution was adjusted by adding 1 M CaCl2 solution filtered with a 0.45 μm filter. Disposable
plastic cuvettes were used for DLS measurements, repeated three times
with 25 runs per measurement.
ζ
Potential and Electrophoretic Mobility
Studies
The effect of Ca2+ ion concentration on
electrophoretic mobility and the ζ potential of CMC solutions
and CNF (in suspension) were analyzed using Zetasizer Nano Zs (Malvern
instruments U.K.). A 0.5% (w/v) CNF suspension was probe sonicated
in an ice bath for 1 min and diluted 10 times to get a CNF suspension
of concentration 0.05% (w/v); the electrolyte concentration of the
suspension was adjusted by adding an appropriated amount of 1 M CaCl2. The CMC solutions prepared for DLS measurements were used
for electrophoretic mobility measurements.
Adsorption
Experiments Using QCM-D
The adsorption of CMC onto the cellulose
surface from the solutions
of varying concentrations of CaCl2 ranging from 5 to 250
mM was studied using QCM-D equipment (Biolin Scientific, Gothenburg,
Sweden). All of the experiments were performed at 25 °C with
a flow rate of 20 μL/min. A concentration of 0.2% (w/v) CMC
solutions containing different concentrations of CaCl2 were
injected into flow cells. The CNF film was equilibrated with CaCl2 solutions of corresponding concentrations prior to the injection
of the CMC solution.Similarly, adsorption of CMC on cellulose
at 250 mM ionic strength of CaCl2 has been performed from
D2O and H2O. A model introduced by Johannsmann
et al. was used to calculate the adsorbed mass.[31] According to Johannsmann, the shift in the complex frequency
is related to the resonance frequency of the crystal in solution by
the following equation.where δ̂f is the shift in the
complex frequency, f0 is the fundamental
resonance frequency of the quartz
crystal in air, f is the resonance frequency of the
crystal in contact with the solution, d is the thickness
of the film, and is the complex shear compliance. ρq and μq are the specific density and elastic
shear modulus of the quartz crystal. Equation can be written in a simpler form using equivalent
mass (m*), which is defined asThen, we
obtain a linear equationIt is assumed that ^J(f) is independent of the frequency in the accessible
range and the true sensed mass mo is obtained
graphically by plotting equivalent mass against the square of the
resonance frequency. In this investigation, third, fifth, and seventh
overtones were used for Johannsmann’s modeling. It is important
to mention that the true sensed mass calculated using Johannsmann’s
modeling includes the mass of water that is associated with the adsorbed
layer and thus not equal to the dry mass of adsorbed CMC.
Swelling and Deswelling Studies Using QCM-D
The adsorbed
layers of CMC were subjected to swelling by introducing
distilled water into the flow cell with a flow rate of 20 μL/min.
It is important that the flow rate be kept small to prevent flow-induced
desorption. The swollen CMC layers were subjected to deswelling by
the introduction of calcium chloride solution into the flow cell.
Results and Discussion
Characterization
of the CNF Model Film
The spin coating of CNF on QCM-D sensors
resulted in an ultrathin
film with the fibrillar morphology illustrated in the AFM height image
(Figure ). The film
showed a root-mean square roughness of 3.2 nm, determined using an
area of 5 μm × 5 μm (Figure ).
Figure 1
AFM height image of the CNF model film.
AFM height image of the CNF model film.Immersion of the film into water, followed by solvent
exchange
to D2O, and then using the 10% mass difference of H2O and D2O to determine the water content of the
film[30,32] revealed that the CNF film is highly hydrated
(41 mg m–2, see the Supporting Information S1). The solvent fraction estimated in the film
is considered to be an indirect measure of the void volume on the
surface of the CNF model film.[33] Both AFM
and solvent exchange studies revealed that the CNF film has a morphology
that contributes to a large surface area that is available for interaction
with the solvent and the adsorbates.
Carboxymethylcellulose
in CaCl2 Solutions
The CMC solutions (0.2%w/v)
were visually transparent;
no phase separation could be observed in any of the concentrations
of CaCl2 used in this study (5–250 mM). These results
were contradictory to the observations made by Sharratt et al.,[24] who showed that CMC shows an L-type phase behavior,
i.e., a separate phase in the presence of very small concentrations
of specifically interacting cations such as Ca2+ and Ba2+. The critical concentration of salt required to induce phase
separation depends on the types of CMC (Mw and DS), salt, and the
concentration of CMC.[34−36] The CMC used in the present study has a significantly
lower Mw (95 kDa) and a lower DS than the CMC used by Sharratt et
al. (250 kDa),[24] which can explain the
difference observed in the phase behavior.The hydrodynamic
diameters of the CMC solutions were determined using dynamic light
scattering. The primary particle size distribution (PSD) obtained
from the light scattering experiment is intensity PSD (red curve in Figure a), showing that
the majority of the CMC has a hydrodynamic diameter exceeding 100
nm. In contrast, the volume PSD shows a single sharp peak, indicating
that the majority of the CMC volume has a hydrodynamic diameter within
5 nm (Figure b). The
contrasting information of the hydrodynamic size obtained from the
intensity PSD and volume PSD of the same solution is due to the method
used for extracting the hydrodynamic diameter from the autocorrelation
function. The calculation of intensity PSD relies solely on the intensity
of the light scattered by the solution. The larger particles dominate
the scattering and, accordingly, they affect the particle size distribution
more. Mie theory was used in the calculation of volume PSD, where
optical properties such as the refractive index and the absorption
of light by the material are considered together with the scattered
intensity.[37] Only the volume PSD will be
used in the discussion henceforth since this is the standard procedure
for reporting hydrodynamic size.
Figure 2
(a) Volume (red) and intensity (blue)
PSD of CMC in 5 mM CaCl2. (b) Volume PSDs of the CMC for
different ionic strengths
of CaCl2.
(a) Volume (red) and intensity (blue)
PSD of CMC in 5 mM CaCl2. (b) Volume PSDs of the CMC for
different ionic strengths
of CaCl2.Figure b represents
the volume PSD of CMC solutions in 5–250 mM concentration of
CaCl2. The increase in the concentration of Ca2+ ions in the solutions resulted in a larger average size and wider
size distributions. The average size obtained from the major peak
in the volume PSD is plotted as a function of the CaCl2 concentration (Figure a).
Figure 3
(a) Average hydrodynamic diameter obtained from the volume PSD
and (b) electrophoretic mobility of CMC solutions for different ionic
strengths of CaCl2.
(a) Average hydrodynamic diameter obtained from the volume PSD
and (b) electrophoretic mobility of CMC solutions for different ionic
strengths of CaCl2.A linear increase in hydrodynamic size can be observed from 5 to
100 mM CaCl2. The curve levels off after 100 mM and, at
250 mM CaCl2, the average hydrodynamic diameter was around
25 nm. This concurs with the observation of Sharratt et al.[24]The reason for the trend observed in the
hydrodynamic diameter
can be attributed to the interaction of Ca2+ with CMC polymer
chains. It has been suggested that Ca2+ ions interact with
CMC chains both through ion–ion correlations and dispersion
interactions.[38] This could reduce the charge
density of CMCs close to neutral, as is evident from the electrophoretic
mobility studies shown in Figure b. The electrophoretic mobility shifted to less negative
values as the concentration of CaCl2 increased and started
to level off from 60 mM. The reduction in the charge of CMC chains
can promote both inter- and intrachain associations and the formation
of multichain clusters, thereby leading to the trend observed for
the hydrodynamic diameter of the CMC in different CaCl2 concentrations.
CNF Suspensions in Different
Concentrations
of CaCl2
The effect of the concentration of CaCl2 on a cellulose surface was studied by analyzing the ζ
potential of CNF suspensions (Figure ). Ideally, this should have been done on CNF-coated
QCM-D sensors: due to the experimental limitations of this study,
it was performed on CNF suspensions instead. In CNF suspensions, there
is a possibility of structural reorganization of CNFs in the presence
of divalent cations. Valencia et al. have reported re-entrant transitions
in the microstructure of TEMPO-oxidized CNF suspensions in the presence
of divalent ions.[39] A similar reorganization
caused by cations on the CNF microstructure within a film might be
more limited compared to observed for dispersions.
Figure 4
ζ Potential of
CNF suspensions in different ionic strengths
of CaCl2.
ζ Potential of
CNF suspensions in different ionic strengths
of CaCl2.The CNF suspensions showed
a ζ potential of −35 mV
before adding CaCl2, indicating that the CNF surface was
negatively charged and the suspension colloidally stable.[40,41] The addition of 5 mM CaCl2 destabilized the colloidal
suspension of CNF and resulted in a slightly turbid suspension. It
should be kept in mind that the CNF suspensions were not colloidally
stable in all of the concentrations of CaCl2 used: the
ζ potential values obtained and presented in Figure , therefore, manifest not only
the ζ potential but also the instability and should therefore
only be considered as being qualitative in later discussions. However,
the precipitation observed of CNFs indicates that surface charges
are screened to a large extent, even at low concentrations of CaCl2.
Adsorption of CMC from CaCl2 Solutions
The QCM-D technique enables real-time investigation of the adsorption
process to be made in terms of the kinetics of adsorption, the viscoelastic
nature of the adsorbed layers, and the mass adsorbed per unit area. Figure shows representative
QCM frequency and dissipation curves for CMC adsorption in the presence
of CaCl2.
Figure 5
Representative (a) frequency (third overtone) and (b)
dissipation
curves for the adsorption of CMC on CNF model surfaces in aqueous
CaCl2 environments.
Representative (a) frequency (third overtone) and (b)
dissipation
curves for the adsorption of CMC on CNF model surfaces in aqueous
CaCl2 environments.Some general conclusions can be drawn from these adsorption curves:
first, the adsorption process of CMCs onto the CNF model film is slow
for all concentrations of CaCl2 studied, with the adsorption
not reaching a plateau even after 90 min. The change in frequency
and dissipation shift increase with the increasing concentrations
of CaCl2, clearly showing that the CaCl2 concentration
has a profound effect on the CMC adsorption process. This is in accordance
with observations made by Laine et al.[4] and Liu et al.,[8] who studied the adsorption
of CMC onto cellulose-rich macrofibers and regenerated cellulose surfaces,
respectively. The high dissipation values indicate that the adsorbed
layer is viscoelastic. Further structural information on adsorbed
CMC layers can be obtained by analyzing the so-called D–f plots,
which represent the changes in the conformation of the adsorbed layer
qualitatively as adsorption proceeds. Figure shows representative D–f plots of
the adsorption of CMC in different concentrations of CaCl2.
Figure 6
Change
in dissipation as a function of normalized frequency shift
(third overtone) during the adsorption of CMC on CNF model surfaces
in aqueous CaCl2 environments.
Change
in dissipation as a function of normalized frequency shift
(third overtone) during the adsorption of CMC on CNF model surfaces
in aqueous CaCl2 environments.D–f plots of CMC adsorption show a continuous curve in all
of the CaCl2 concentrations studied, further confirming
that the adsorption of CMC is a slow process.[42] The increase in the slope of the D–f curves at lower concentrations
of CaCl2 suggests that the adsorbed layer is more viscous,
which is indicated by the increase in energy dissipation. An increase
in the CaCl2 concentration, on the other hand, decreases
the slope and hence the viscous component, resulting in the adsorbed
CMC layer becoming denser.The degree of collapse of the polymer
structure upon adsorption
depends on the conformation of the polymers in solution (Figure a) and the strength
of the interaction of the polymer with the surface.[43] The denser layer formed at higher concentrations of CaCl2 indicates that the structure of polymer molecules that come
into contact with the surface collapse (Figure b), which is also an indication of attractive
interactions between the CMC and the cellulose surface. Kargl et al.
suggested that CMC has a specific interaction with a cellulose surface,[7] although the nature of this interaction has not
yet been identified. The conformational rearrangement and spreading
over the surface could be hampered kinetically by the adsorption of
neighboring molecules. Consequently, some of the adsorbed CMC might
not be directly involved in the immediate interaction with the cellulose
surface (Figure c),
which contributes to increasing the viscosity of the layer and hence
the dissipation of energy. As the adsorption proceeds, the D–f
curves (Figure ) show
a nonlinear adsorption behavior with a slope change, revealing that
the conformation of the adsorbed CMC layer changes during adsorption.
Similar observations have been reported by Köhnke et al. with
respect to the adsorption of arabinoxylans on cellulose.[42] The lower slope of region B in the D–f
plot (Figure ) indicates
that dangling CMC layers diffuse toward the surface and transform
the adsorbed layer, making them denser. It could also be arising due
to conformational changes in the adsorbed layer due to the exchange
of shorter adsorbed chains with larger CMC chains, which is thermodynamically
favored.
Figure 7
Schematic representation of CMC adsorption in the presence of CaCl2 (orange dots). (a) CMC chains (green) approach the surface.
(b) Collapse and spread of CMC chains on the cellulose surface and
into the pores of the CNF film. (c) Dangling CMC chains interact with
adsorbed CMC on the cellulose surface (second stage) and desorption
of shorter chains. (d) Dangling CMC chains penetrate the adsorbed
CMC network to form a thicker adsorbed layer.
Schematic representation of CMC adsorption in the presence of CaCl2 (orange dots). (a) CMC chains (green) approach the surface.
(b) Collapse and spread of CMC chains on the cellulose surface and
into the pores of the CNF film. (c) Dangling CMC chains interact with
adsorbed CMC on the cellulose surface (second stage) and desorption
of shorter chains. (d) Dangling CMC chains penetrate the adsorbed
CMC network to form a thicker adsorbed layer.It is possible that Ca2+ ions can still induce multichain
association on the surface and, at this stage, the adsorbed layer
can be considered as being a Ca2+ cross-linked gel network
that has been attached to the cellulose surface.The swelling
and deswelling behavior of the adsorbed CMC layers
in the presence of deionized water and 20 mM CaCl2 is illustrated
in Figure . An immediate
drop in frequency is observed when 20 mM CaCl2 is exchanged
with deionized water. This corresponds to the water intake, which
can also be termed as swelling of the adsorbed layer. Further introduction
of 20 mM CaCl2 into the flow cell at 40 and 68 min increased
in frequency, showing that the CMC layer deswells or collapses in
the presence of Ca2+. Similar increases in frequency upon
water injection into the CMC adsorbed film have been reported by Kargl
et al.[7] The swelling–deswelling
behavior was reversible, two cycles of which are presented in Figure , thereby confirming
the gel network nature of the adsorbed layers.
Figure 8
Swelling and deswelling
cycles of the adsorbed layers of CMC on
a cellulose surface.
Swelling and deswelling
cycles of the adsorbed layers of CMC on
a cellulose surface.The adsorbed CMC per
unit area, calculated using Johannsmann’s
model, is plotted against the concentration of CaCl2, as
shown in Figure .
Figure 9
Mass of
the adsorbed CMC per unit area on a cellulose surface from
CMC solutions with different concentrations of CaCl2.
Mass of
the adsorbed CMC per unit area on a cellulose surface from
CMC solutions with different concentrations of CaCl2.Adsorption increases as the concentration of CaCl2 increases,
which agrees with the observations of Laine et al. and Liu et al.[4,8] The mass of the CMC adsorbed per unit area increased significantly
from 5 to 60 mM CaCl2 and started to level off. The ionic
strength 5 mM corresponds to a Ca2+ to CMC charge ratio
of 4. (The properties of the CMC solution has also been expressed
in terms of the charge ratio between Ca2+ and CMC, which
can be found in the Supporting Information.) As can be seen in Figure , the shape of this adsorption isotherm is somewhat similar
to the isotherm for polyelectrolytes adsorbed on like-charged surfaces,
as predicted by Scheutjens–Fleer polymer adsorption theory
(S-F theory). In this theory, polymer adsorption is described in terms
of four parameters: χ, ,, and σ, where χ and χ are the Flory–Huggins parameters accounting
for polymer–solvent interactions and the polymer segment-surface
interaction, respectively.[44] Both χ and χ have
a positive linear dependence on the adsorbed amount. The terms and σ represent the surface charge and polymer
segmental charge, respectively, and account for Columbic interactions
in polyelectrolyte adsorption.[44−46] According to S-F theory, the
charges present in both the polyelectrolytes and the surface at higher
salt concentrations are screened (i.e., and σ), and the adsorption will be purely governed by nonelectrostatic
forces, which is accounted for by χ. It is noteworthy that S-F theory only considers monovalent
salts. However, a comparison of the ability of divalent ions with
monovalent ions to increase the adsorbed amount at the same ionic
strength suggests that, apart from obvious electrostatic screening
effects, divalent metal ions contribute positively to the adsorption
process.[8] This positive contribution of
multivalent ions to adsorption might originate from the structural
changes in polymer solutions induced by multivalent ions.[25,47]Figure a,b
shows
how the adsorbed CMC is related to the electrophoretic mobility and
structural properties of the CMC in the presence of CaCl2, respectively. The mass of CMC adsorbed per unit area on cellulose
increases with increases in both electrophoretic mobility and hydrodynamic
diameter. However, when combining data in Figures and 9, no such correlation
could be observed between the ζ potential of the CNF surface
and the adsorbed CMC (data not shown). From the observed ζ potential
values, it is evident that even at an ionic strength of 10 mM, the
charge on the surface becomes close to neutral. Further increase in
ionic strength did not change the surface charge of the CNF appreciably.
On the other hand, the CMC adsorbed per unit area increased with increased
ionic strength up to 250 mM, meaning that Ca2+ ions have
a profound effect on the solution side, which contributed to the adsorption.
This explains the correlation between the CMC adsorbed and the hydrodynamic
size. Moreover, screening negative charges and multichain association
of CMC in the presence of CaCl2 reduces the polymer–solvent
interaction, which could contribute to the driving force of adsorption.
However, this might not be the sole driver of the adsorption of CMC
onto cellulose. Recently, it has been suggested that the adsorption
of hemicellulose on highly hydrated systems such as cellulose is driven
by the entropy gain due to the release of structured water.[26,27] Furthermore, a similar mechanism has been suggested for the adsorption
of negatively charged PEDOT: PSS on the cellulose surface.[28] To test this hypothesis in the case of CMC adsorption,
we compared the adsorption of CMC from deionized water and deuterated
water (D2O) at ionic strength 250 mM CaCl2,
where the adsorption is expected to be driven by nonelectrostatic
forces, according to Fleer et al.[44]
Figure 10
(a) Adsorbed
CMC per unit area at different ionic strengths of
CaCl2 plotted against electrophoretic mobility. (b) Hydrodynamic
dynamic diameter at a corresponding concentration of CaCl2.
(a) Adsorbed
CMC per unit area at different ionic strengths of
CaCl2 plotted against electrophoretic mobility. (b) Hydrodynamic
dynamic diameter at a corresponding concentration of CaCl2.The observed change in frequency
shift indicates that CMC adsorption
is more favored from D2O than H2O. Thus, even
though D2O and H2O are often considered similar
liquids, there is a significant difference in their physicochemical
properties stemming from differences in their intermolecular forces.[48] D2O forms a stronger and higher average
number of hydrogen bonds (10% more) compared to water.[49−51] Thus, it is a more structured liquid than H2O.[52] Therefore, when CMC is adsorbed from D2O, the entropy gain due to the release of structured D2O molecules from the cellulose surface would be more significant
and result in high adsorption than CMC adsorbed from H2O, as is seen in Figure . It has also been suggested that in D2O, hydrophobic
interactions are more significant.[53] Seminal
work by Laine et al. has shown that the adsorption of CMC increases
with an increase in temperature, which also supports the claim of
the entropy-driven mechanism of CMC adsorption.[4,27]
Figure 11
(a)
Representative QCM-D frequency curve of adsorption of CMC on
cellulose from D2O and H2O. (b) Adsorbed CMC
calculated using Johannsmann’s model.
(a)
Representative QCM-D frequency curve of adsorption of CMC on
cellulose from D2O and H2O. (b) Adsorbed CMC
calculated using Johannsmann’s model.
Conclusions
Previous investigations of the
adsorption of CMC on cellulose,
negatively charged polyelectrolytes in general, have overlooked the
changes that occur in the properties of the polymer solution in the
presence of multivalent ions, such as Ca2+. In this study,
the effect divalent ions have on the adsorption of CMC onto cellulose
has been studied from the perspectives that the adsorption process
is multifaceted and can be influenced by the characteristics of the
cellulose surface and the solution. Dynamic light scattering experiments
performed on CMC solutions with varying concentrations of CaCl2 revealed the multichain association of CMC in the presence
of CaCl2, which increased the hydrodynamic size. Furthermore,
a correlation has been observed between the hydrodynamic diameter
of CMCs in different concentrations of CaCl2 and the amount
of CMC adsorbed in the corresponding concentrations of CaCl2. This suggests that the adsorption of CMCs at high concentrations
of CaCl2 is mainly governed by structural changes in the
CMC, which effectively reduce polymer–solvent interactions
and thus contribute to the driving force of adsorption. Furthermore,
we provided an experimental evidence for the entropy-driven adsorption
of CMC on cellulose by comparing the adsorption of CMC from D2O and H2O.
Authors: Joshua D Kittle; Xiaosong Du; Feng Jiang; Chen Qian; Thomas Heinze; Maren Roman; Alan R Esker Journal: Biomacromolecules Date: 2011-06-24 Impact factor: 6.988
Authors: Tobias Benselfelt; Emily D Cranston; Sedat Ondaral; Erik Johansson; Harry Brumer; Mark W Rutland; Lars Wågberg Journal: Biomacromolecules Date: 2016-08-17 Impact factor: 6.988