Effects of pH and Salt Concentration on the Formation andProperties of Chitosan−Cellulose Nanocrystal Polyelectrolyte−Macroion ComplexesHezhong Wang,†,‡ Chen Qian,†,§ and Maren Roman*,†,‡
†Macromolecules and Interfaces Institute, ‡Department of Wood Science and Forest Products, and §Department of Chemistry,Virginia Tech, Blacksburg, Virginia 24061, United States
*S Supporting Information
ABSTRACT: This study examines the effects of pH and saltconcentration on the formation and properties of chitosan−cellulose nanocrystal (CNC) polyelectrolyte−macroion com-plexes (PMCs). The components’ pK values, determined bypotentiometric titration, were 6.40 for chitosan and 2.46 forthe CNCs. The turbidity of PMC particle suspensions wasmeasured as a function of chitosan−CNC ratio, pH, and saltconcentration. The maximum turbidity values in titrations of achitosan solution with a CNC suspension and vice versaoccurred at charge ratios of 0.47 ± 0.11 (SO3
−/NH3+) and 1.16
± 0.06 (NH3+/SO3
−), respectively. A pH increase caused a turbidity decrease due to shrinking of the PMC particles uponchanges in their components’ degrees of ionization. An increase in salt concentration caused a decrease in turbidity due tocharge-screening-related shrinking of the PMC particles. The effects of pH and salt concentration on particle size were confirmedby scanning electron microscopy.
■ INTRODUCTIONPolyelectrolytes are polymers with ionic or ionizable groups.Upon mixing, polyelectrolytes of opposite charge formpolyelectrolyte complexes (PECs) under release of thecounterions to the surrounding medium. The properties ofPECs depend on a number of factors, predominantly thecomponents’ molecular weights, densities of ionic groups,mixing ratio, concentrations upon mixing, and the ionicstrength of the surrounding medium. For polyelectrolyteswith acidic or basic groups, the pH of the surrounding mediumis another important factor because it affects their degrees ofionization and thus charge densities. Several studies have beenperformed to analyze the effects of polyelectrolyte chargedensity on the formation, structure, and composition ofPECs.1−8 The studies have shown that polyelectrolytes withsimilar charge densities form densely structured PECs whereaspolyelectrolytes with dissimilar charge densities form highlyswellable PECs with lower structural densities.5,8 Thesefindings are in agreement with the results of numerous studieson the pH-sensitive swelling behavior of PECs involvingpolyacids and polybases.9−16 PEC swelling upon changes in pHis due to changes in the charge stoichiometry of the PEC,resulting from changes in the degrees of ionization of thepolyelectrolytes. PECs with a stoichiometric charge ratio aremore densely structured than PECs with a nonstoichiometriccharge ratio, but because they have no net charge they areprone to aggregation, whereas PECs with a nonstoichiometriccharge ratio are charge stabilized. The effects of low-molecular-weight electrolytes on the formation and stability of PECs,
which are due to the ability of salt ions to screen the Coulombinteractions between the two polyelectrolytes, have also beenstudied extensively.17−32 Small amounts of salt, present duringPEC formation, enable rearrangement processes toward a stateof thermodynamic equilibrium.28 Addition of small amounts ofsalt subsequent to PEC formation causes PECs to shrinkbecause of charge screening. Higher salt concentrations enablerearrangement and flocculation of PECs through secondaryaggregation.24 Depending on the types of ionizable groups ofthe polyelectrolytes, very high salt concentrations may result inthe dissociation of PECs.30−32
We have recently reported a new type of ionic complexbetween chitosan, a linear copolysaccharide of β(1−4)-linked2-amino-2-deoxy- and 2-acetamido-2-deoxy-D-glucopyranoseresidues, and cellulose nanocrystals (CNCs), cylindricalnanoparticles of cellulose, for potential use as a multiparticulateoral drug delivery system.33 Chitosan is positively charged inacidic media because of the protonation of its amino groups.CNCs are negatively charged at neutral and basic pH valuesbecause of sulfate groups on their surface, resulting from partialesterification of surface hydroxyl groups during their preparationby sulfuric acid hydrolysis. On the basis of the terminology usedin the theoretical literature, we termed these ionic complexespolyelectrolyte−macroion complexes (PMCs). Our previousreport focused on the effects of chitosan concentration, the
Received: July 13, 2011Revised: September 4, 2011Published: September 21, 2011
components’ mixing ratio, and the mixing sequence on theformation and properties of these PMCs. Because of the strongmismatch in the ionizable group densities of the components,the PMC particles were composed primarily of CNCs. Theirsize ranged from a few hundred nanometers to severalmicrometers and depended on the cellulose/chitosan ratio.Under certain conditions, the PMC particles had a peculiarhexagonal shape, which was attributed to the cylindrical shape ofthe CNCs.This study is a continuation of the previous work with the
aim to deepen our understanding of the factors that affect theproperties of chitosan−CNC PMCs with regard to oral drugdelivery applications. Because the gastrointestinal tract has arange of pH levels and salt concentrations, the specificobjectives of this study were to determine the effects of thepH and salt concentration on the formation and properties ofthe PMCs.
Chemika) was purchased from Sigma-Aldrich and purified as follows.Typically, 1 g of chitosan was dissolved overnight in 250 mL of 0.1 NHCl, and the solution was filtered through a series of Milliporepoly(vinylidene fluoride) (PVDF) syringe filters (pore sizes 1, 0.45,and 0.22 μm). Next, chitosan was precipitated by the addition of 1 NNaOH until the solution pH reached 9−10. The purified chitosan wascollected by centrifugation (4900 rpm for 15 min at 4 °C), washedthree times with deionized water, and freeze-dried overnight. Thepurified chitosan was characterized as described previously.33 Theviscosity-average molecular weight and degree of deacetylation were2.4 × 105 g/mol and 88%, respectively. The amino group density was5.83 mol/kg.
CNCs were prepared and characterized as described previously.33 Inbrief, 50 g of ground (60-mesh) dissolving-grade softwood sulfite pulp(Temalfa 93 A-A), kindly provided by Tembec, Inc., was treated with500 mL of 64 wt % H2SO4 at 45 °C for 45−60 min. The hydrolysiswas stopped by 10-fold dilution of the reaction medium with deionizedwater. The CNCs were collected by centrifugation and dialyzed againstdeionized water until the pH of the dialysis water stayed constant. Theobtained suspension (∼250 mL) was sonicated under ice-bath coolingfor 15 min at 40% output with a 500 W ultrasonic processor (Sonics &Materials, model VC-505), equipped with a 13 mm (tip diameter)titanium alloy probe, and subsequently filtered through a 0.45 μm andthen 0.22 μm PVDF syringe filter. The concentrations of the filteredCNC stock suspensions (three different batches) were 0.76, 0.78, and0.88% (w/v). The CNCs in the three batches, being in the acid form,had sulfate group densities of 0.38, 0.18, and 0.33 mol/kg, respectively,determined in triplicate by conductometric titration.33
H2SO4 (>95%), HCl (0.1 and 5 N, certified), NaOH (0.1 and 1 N,certified), and NaCl (certified) were purchased from Fisher Scientific.The water used in the experiments was deionized water from aMillipore Direct-Q 5 ultrapure water system (resistivity at 25 °C: 18.2MΩ·cm).Preparation of Chitosan Solutions. Chitosan solutions for the
complexation experiments were prepared from a stock solution of∼0.1% (w/v) by dilution with deionized water. For preparation of thestock solution, purified chitosan was dried in an oven at 105 °C for2 h. Then, 0.1 g of the oven-dried, purified chitosan was dissolved in100 mL of 0.1 N HCl. The exact concentration of the stock solutionwas determined in triplicate by thermogravimetric analysis as describedpreviously.33 The pH and salt concentration of the dilute chitosansolutions were adjusted with 0.1 N HCl or 0.1 N NaOH and 5 MNaCl, respectively.Preparation of CNC Suspensions. Dilute CNC suspensions for
the complexation experiments were prepared from the filtered stocksuspension by dilution with deionized water. The pH and saltconcentration of the dilute CNC suspensions were adjusted with 0.1 NHCl or 0.1 N NaOH and 5 M NaCl, respectively.
Potentiometric Titration. Potentiometric titration curves weremeasured in triplicate (minimum) with a Mettler Toledo SevenMultiS47 pH/conductivity meter with an InLab 413 pH electrode. In thecase of CNCs, titrations were carried out with 25−50 mL aliquots ofthe filtered stock suspension, adjusted to a salt concentration of 1 mMthrough the addition of 5 M NaCl. In the case of chitosan, the titrandwas 100 mL of a 0.1% (w/v) solution of chitosan in 0.025 N HCl witha salt concentration of 0.1 M. The titrant, 0.02 N NaOH, prepared bydilution of 0.1 N NaOH with deionized water, was added undernitrogen and stirring in 0.5 mL increments. After each addition, thepH of the titrand was recorded. Titration simulations and multipleregression analysis of titration data were performed with CurTiPot forMicrosoft Excel freeware (v3.4.1).Turbidimetric Titration. Turbidimetric titrations were performed
with a 0.001% (w/v) chitosan solution and a 0.02% (w/v) CNCsuspension (sulfate group density: 0.18 mol/kg), both having the samepH and salt concentration. The chitosan and CNC concentrationswere chosen on the basis of the results of our previous study.33 In aturbidimetric titration experiment, the CNC suspension was addeddropwise under vigorous stirring with a magnetic bar to the chitosansolution or vice versa. The transmittance of the reaction mixture wasmonitored with a Brinkmann PC 950 probe colorimeter with a 1 cmoptical cell (2 cm path length), operating at a wavelength of 420 nm.Turbidity values were calculated as 100 − transmittance (%). Prior toeach titration, the probe colorimeter was zeroed in deionized water.Each experiment was performed in triplicate.Characterization of PMC Particles. The morphology of the
PMC particles was analyzed by field-emission scanning electronmicroscopy (FE-SEM). Images were recorded with a LEO 1550 FEscanning electron microscope at an accelerating voltage of 1 kV and aworking distance of 4−5 mm. Particles for SEM imaging wereprepared at an amino/sulfate group molar ratio (N/S ratio) of unity byaddition of a 0.02% (w/v) CNC suspension under vigorous stirringwith a magnetic bar to a 0.001% (w/v) chitosan solution, both havinga pH of 2.6 and a salt concentration of 1 mM. The pH or saltconcentration of the PMC particle suspension was subsequentlyadjusted with 0.1 N HCl or 0.1 N NaOH and 5 M NaCl, respectively,and the particles were allowed to equilibrate for a minimum of 10 min.For SEM sample preparation, a 10 μL drop of the PMC particlesuspension was deposited onto Ni−Cu conductive tape (Ted Pella)mounted onto a standard SEM stub (Ted Pella) and allowed to dryunder ambient conditions. For the highly acidic sample (pH 1), aspherical piece of gold foil was used as the substrate instead of the Ni−Cu tape, because with the Ni−Cu tape, NiCl2 formation interferedwith the analysis. Prior to imaging, the SEM samples were coated witha thin (6 nm) layer of carbon.
■ RESULTS AND DISCUSSION
Effect of pH. The effect of pH on the complexation ofchitosan and CNCs was analyzed by turbidimetric titration atdifferent pH levels. Figure 1 shows the titration curves for thetitration of a chitosan solution with a CNC suspension (type 1)and the reverse direction, i.e., a titration of a CNC suspensionwith a chitosan solution (type 2), at a salt concentration of1 mM and pH levels of 2.6, 3.5, 4.5, and 5.6. In accordance withpreviously reported results,33 in type 1 titrations (Figure 1a),the turbidity of the reaction mixture asymptotically approacheda maximum value with increasing sulfate/amino group molarratio (S/N ratio), whereas the turbidity reached a maximumand then decreased with increasing N/S ratio in type 2titrations (Figure 1b). For both methods, an increase in pHfrom 2.6 to 5.6 resulted in a decrease in maximum turbidity.Furthermore, in type 2 titrations, an increase in pH caused ashift of the turbidity maximum to higher N/S ratios.The turbidity of a colloidal dispersion is a complicated
function of the number per unit volume (number density), size,and optical properties of the light-scattering bodies. In the
system studied here, the number density and size of the light-scattering PMC particles are inversely related. Furthermore, theoptical properties of the PMC particles depend on the opticalproperties of the two components and the particle composition.The particle composition, in turn, depends on the chargestoichiometry in the reaction mixture, which is governed by themass ratio of the two components, their ionizable groupdensities, and their degrees of ionization, i.e., the fractions ofionizable groups ionized. To better understand the compositionof the PMC particles at the four pH values investigated, wedetermined the degree of ionization of each component as afunction of pH.For basic groups, the degree of ionization, αb, is related to the
pH by
(1)
where pKb is the negative decadic logarithm of the dissociationconstant, Kb, of the basic groups. For acidic groups, the degreeof ionization, αa, is related to the pH by
(2)
where pKa is the negative decadic logarithm of the dissociationconstant, Ka, of the acidic groups. The pKb of chitosan wasmeasured by potentiometric titration. A typical potentiometrictitration curve for chitosan is shown in Figure 2. The obtained
pKb, given by the pH at the halfway point between the twoequivalence points, was 6.40 ± 0.03, in good agreement withthe literature.34
Determination of the pKa of the CNCs was more difficultbecause of their low sulfate group density (0.18−0.38 mol/kg).A titration simulation, shown in Figure S1, assuming a pKa of2.46 for the sulfate groups, based on the results of our indirectapproach described below, showed that a barely measurableequivalence point for the onset of sulfate group dissociationwould require 13 g of CNCs with a sulfate group density of0.38 mol/kg suspended in 25 mL of 0.1 N HCl to be titratedwith 2 N NaOH. Under the initial conditions (pH of 0.97,CNC concentration of 53% (w/v)), the CNC suspensionwould be a viscous gel and not titratable. In lack of a practicalmethod for measuring the pKa of the sulfate groups directly, weused the approach of Ikeda et al.15 to determine the degree ofionization of the sulfate groups at different pH values from thecharge balance. The degree of ionization of the sulfate groups,αa, for a direct titration of an aqueous CNC suspension,containing dissolved CO2, with NaOH can be expressed as
(3)
where [H+] is the concentration of hydrogen ions; C, theconcentration of the titrant; V, the volume of the titrant added;V0, the initial volume of the titrand; KW, the ionic product ofwater; Ca, the initial concentration of sulfate groups; CCO2
, theinitial concentration of dissolved CO2; and αCO2
, the fraction ofCO2 molecules present as HCO3
−. αCO2is given by (([H+]/
Ka,CO2) + 1)−1, where Ka,CO2
is the dissociation constant for thereaction CO2 + H2O ⇌ HCO3
− + H+, taken as 6.352.35 Thereaction HCO3
−⇌ CO32− + H+ has been disregarded because it
occurs at pH levels outside the relevant range.A typical potentiometric titration curve for CNCs is shown in
Figure 2. The equivalence point corresponds to the completionof the dissociation of sulfate groups. The dashed line indicatesthe hypothetical shape of the titration curve in the absence of
Figure 1. Turbidimetric titration curves for (a) the titration of achitosan solution with a CNC suspension (type 1) and (b) thetitration of a CNC suspension with a chitosan solution (type 2) fordifferent pH values at a salt concentration of 1 mM. (Data points aremeans of three measurements. Error bars are omitted for clarity. Thedashed lines are guides to the eye.)
Figure 2. Potentiometric titration curves for chitosan (●) and CNCs(○). The dashed line indicates the hypothetical shape of the CNCtitration curve in the absence of dissolved CO2. The dotted linesindicate the equivalence points and halfway point of the chitosantitration curve.
dissolved CO2. Figure 3 shows the degree of ionization valuesobtained with eq 3 from the titration data. The plotted valuesassume a dissolved CO2 concentration of 4% of the total initialconcentration of anion forming species, estimated by multipleregression analysis of the potentiometric titration data (FigureS2). The dashed line in Figure 3 is a fit of eq 2 to theexperimental data using pKa as the fitting parameter. Analysis ofsix titration curves from three different batches of CNCs, withdissolved CO2 concentrations ranging from 1 to 13% of thetotal initial concentration of anion forming species, gave a meanpKa of 2.46 ± 0.12. A slighly higher value (2.8) has beenreported for κ-carrageenan,36 a sulfated galactan, naturallyoccurring in red seaweed. Also shown in Figure 3 is the degreeof ionization curve for chitosan obtained with eq 1 and themeasured pKb of 6.40. It is apparent in Figure 3 that the pHrange in which both components are completely ionized isnarrow and centered at 4.5. At higher pH values, chitosan isincompletely ionized, and at lower pH values, the CNCs losetheir charge.The degrees of ionization of the two components for the
investigated pH values are listed in Table 1. As seen in Table 1,
with increasing pH, the degree of ionization of the aminogroups decreased and that of the sulfate groups increased,causing a decrease in NH3
+/SO3− charge ratio with increasing
pH at any given N/S ratio. Consequently, at higher pH values,an NH3
+/SO3− charge ratio of unity required a higher N/S ratio
and lower cellulose/chitosan mass ratio. The cellulose/chitosanmass ratio for an NH3
+/SO3− charge ratio of unity decreased
from 57 at a pH of 2.6 to 28 at a pH of 5.6. (Previously, we
reported a degree of ionization of 0.5 for the CNCs at pH 2.6.33
That value was based on the pKa value from a singlepotentiometric titration curve. The value reported here (0.58)is based on the mean pKa from six titration curves and istherefore probably more accurate.)The decrease in maximum turbidity in the turbidimetric
titration curves (Figure 1) with increasing pH was thereforeattributed to the decrease in NH3
+/SO3− charge ratio of the two
components, requiring fewer CNCs at higher pH values tocompensate the charge of chitosan. The decrease in NH3
+/SO3−
charge ratio with increasing pH was most likely also the reasonfor the shift of the turbidity maximum to higher N/S ratios inthe type 2 titrations (Figure 1b) because it meant that morechitosan was needed for charge neutralization in the PMCs. Asmentioned above for PECs, a net charge of zero allows particleaggregation, resulting in increased turbidity levels. The decreasein turbidity after the maximum has previously been attributedto a reversion of PMC aggregation due to charge over-compensation by incorporation of additional chitosan mole-cules into the PMCs, resulting in a net positive charge andcharge stabilization of the PMCs.33 The curve maxima at pH2.6, 3.5, 4.5, and 5.6 occurred at the N/S ratios 0.7, 1.0, 1.2, and1.3, respectively, corresponding to an NH3
+/SO3− charge ratio
of 1.16 ± 0.06. In the type 1 titrations (Figure 1a), themaximum turbidity levels at pH 2.6, 3.5, 4.5, and 5.6 werereached at the S/N ratios 0.3, 0.5, 0.7, and 0.7, correspondingto an SO3
−/NH3+ charge ratio of 0.47 ± 0.11.
To elucidate the effect of pH on the morphology ofchitosan−CNC complexes, we studied the turbidity of a PMCparticle suspension with an N/S ratio of unity, prepared byaddition of a CNC suspension to a chitosan solution, as afunction of pH. The initial pH and salt concentration of thePMC particle suspension were 2.6 and 1 mM, respectively. Theturbidity of the suspension was monitored upon addition of 0.1N NaOH or 5 N HCl. Figure 4 shows the changes in turbidity
of the reaction mixture with increasing or decreasing pH, withrespect to the starting pH of 2.6. The degree of ionizationcurves are overlaid for reference. Lowering of the reactionmixture’s pH from 2.6 to 0.9 caused a strong increase inturbidity. As seen from the degree of ionization curves, thedecrease in pH from 2.6 to 0.9 resulted in a decrease in thedegree of ionization of the CNCs from 0.58 to 0.03, whereas
Figure 3. Degree of ionization as a function of pH for chitosan (αb)and CNCs (αa): () αb for pKb = 6.40 (measured), (○) experimentalαa values, (−−) fit of eq 2 to the experimental data.
Table 1. Degrees of Ionization of the Chitosan AminoGroups, αb, and the CNC Sulfate Groups, αa; NH3
+/SO3−
Charge Ratio in the Reaction Mixture at an N/S Ratio ofUnity; and Cellulose/Chitosan Mass Ratio in the ReactionMixture at an NH3
Figure 4. Turbidity of a PMC particle suspension with a saltconcentration of 1 mM and an N/S ratio of unity as a function of pH.(The dashed line is a guide to the eye. The degree of ionization curvesare overlaid for reference.)
the degree of ionization of chitosan stayed constant at a value ofunity. The observed increase in turbidity was thereforeattributed to an increase in the size of the PMC particleswith increasing NH3
+/SO3− charge ratio as more and more
CNCs were needed to compensate the charge of a givennumber of ammonium groups. In addition, the chitosan chainsmay partially uncoil as some of the ammonium groups alongthe chitosan backbone lose their counterion (SO3
−). Raising thereaction mixture’s pH from 2.6 to 11.4 caused initially nochange in turbidity and then a decrease above a pH value ofabout 5.5. At pH 5.5, the CNCs had a degree of ionization ofunity and chitosan had a degree of ionization of 0.89. Thedegree of ionization of chitosan reached a value of zero at a pHof about 9. The observed decrease in turbidity was attributed toa decrease in the size of the PMC particles. As the chitosandegree of ionization decreased, the PMC particles may havepartially dissociated and rearranged into smaller particles with ahigher chitosan/cellulose mass ratio. Furthermore, the decreasein chitosan charge density may have caused the chitosanmolecules to form tighter coils and precipitate. The observeddecrease in PMC particle size with increasing pH above 5.5 wasin agreement with the results of Lopez-Leon et al.,37 whoobserved a decrease in mean particle diameter for a pH increasefrom 4 to 7 in suspensions of chitosan−tripolyphosphatenanoparticles.The morphology of chitosan−CNC PMC particles equili-
brated at different pH values were analyzed by FE-SEM.Figure 5 shows SEM images of particles equilibrated at a pH of
1.0, 4.5, 6.5, 7.5, and 9.0. At all pH values, the particles wereroughly spherical in shape. The SEM images confirmed adecrease in particle size with increasing pH.Effect of Salt Concentration. The effect of a low-
molecular-weight electrolyte on the complexation of chitosanand CNCs was studied by turbidimetric titration at differentNaCl concentrations. Figure 6 shows the titration curves for the
titration of a chitosan solution with a CNC suspension (type 1)at a pH of 2.6 and salt concentrations of 0, 0.1, and 0.4 M. Themaximum turbidity observed in the presence of salt was slightlylower than that observed in the absence of salt. Furthermore, inthe absence of salt, the maximum turbidity was reached at anS/N ratio of 0.5, whereas at salt concentrations of 0.1 and 0.4M, the maximum turbidity was reached at S/N ratios of 0.75and unity, respectively. In other words, at S/N ratios belowunity a higher salt concentration resulted in a lower turbidity.As mentioned earlier, salt ions present during the formation ofPECs have been shown to enable rearrangement processestoward a state of thermodynamic equilibrium.28 The lowerturbidity values observed at higher salt concentrations mighttherefore indicate smaller PMC particles with closer-to-equilibrium structures.The effect of salt addition after PMC formation was studied
by turbidity measurements of PMC particle suspensionsequilibrated at different salt concentrations subsequent toPMC formation. Figure 7 shows the turbidity of a PMC particle
suspension with an N/S ratio of unity as a function of saltconcentration. The initial pH and salt concentration of thePMC suspension were 2.6 and 1 mM, respectively. The salt
Figure 5. FE-SEM images of PMC particles equilibrated at differentpH values: (a) 1.0, (b) 4.5, (c) 6.5, (d) 7.5, (e) 9.0. Scale bar: 1 μm(applies to all images).
Figure 6. Turbidimetric titration curves for the titration of a chitosansolution with a CNC suspension (type 1) for different saltconcentrations at a pH of 2.6. (Data points are means of threemeasurements. Error bars represent one standard deviation. Thedashed lines are guides to the eye.)
Figure 7. Turbidity of a PMC particle suspension with a pH 2.6 and areaction mixture N/S ratio of unity as a function of salt concentration.(Data points are means of three measurements. Error bars represent ±one standard deviation. The dashed line is a guide to the eye.)
concentration of six aliquots of the suspension was adjusted todifferent values through addition of 5 M NaCl, and theturbidity of each aliquot was determined after a shortequilibration period (3 min). The turbidity decreased withincreasing salt concentration up to a salt concentration of0.1 M. Upon further increase of the salt concentration, theturbidity leveled off and asymptotically approached a minimumturbidity level.The observed changes in turbidity with increasing salt
concentration (Figure 7) are in accordance with the observedmaximum turbidity levels in the turbidimetric titration curves(Figure 6). The maximum turbidity in the titrations decreasedwith an increase in salt concentration from 0 to 0.1 M andstayed constant with an increase from 0.1 to 0.4 M. Asdemonstrated above, at a pH of 2.6, only 58% of the sulfategroups on the CNCs are charged. Consequently, an N/S ratioof unity represents an NH3
+/SO3− charge ratio of 1.72 and
therefore an excess of chitosan ammonium groups. A low-molecular-weight electrolyte will partially screen the excesscharge and enable the chitosan molecules to take on a moretightly coiled conformation, resulting in a shrinking of theparticles.The size and morphology of chitosan−CNC PMC particles
equilibrated at different salt concentrations were analyzed byFE-SEM. Figure 8 shows SEM images of particles equilibrated
at salt concentrations of 1, 5, 100, and 500 mM. The SEMimages confirm that the particle size decreased with increasingsalt concentration.
■ CONCLUSIONS
This study has shown that the formation and properties ofchitosan−CNC PMC particles are strongly affected by the pH.The effect is based on the influence of the pH on the degrees ofionization and, thus, charge densities of chitosan and CNCs. Adecrease in pH below about 3 causes swelling of the PMCparticles due to a decrease in the charge density of the CNCs.An increase in pH above about 5.5 causes shrinking of the PMCparticles due to a decrease in the charge density of chitosan.The salt concentration seems to have a less pronounced effecton the formation and properties of chitosan−CNC PMCparticles. Higher salt concentrations during PMC formationresult in smaller PMC particles, and addition of salt after PMC
formation causes shrinking of the PMC particles due to chargescreening.
■ ASSOCIATED CONTENT*S Supporting InformationSimulated potentiometric titration curve for the direct pKadetermination of CNCs and multiple regression analysis ofpotentiometric titration data. This material is available free ofcharge via the Internet at http://pubs.acs.org.
■ ACKNOWLEDGMENTSThis material is based upon work supported in part by theUSDA/CSREES under Grant 2005-35504-16088, the NationalScience Foundation under Grant CHE-0724126, and theInstitute for Critical Technology and Applied Science atVirginia Tech. Additional support from Omnova, Inc., andTembec, Inc., is also acknowledged. Furthermore, H.W. thanksthe staff of the Nanoscale Characterization and FabricationLaboratory for assistance with the SEM images.
■ REFERENCES(1) Dautzenberg, H.; Linow, K. J.; Philipp, B. Acta Polym. 1982, 33,
619−625.(2) Koetz, J.; Linow, K. J.; Philipp, B.; Li, P. H.; Vogl, O. Polymer
Bunsen-Ges. Phys. Chem. 1996, 100, 1024−1032.(9) Sui, Z. J.; Jaber, J. A.; Schlenoff, J. B. Macromolecules 2006, 39,
8145−8152.(10) Sakiyama, T.; Takata, H.; Kikuchi, M.; Nakanishi, K. J. Appl.
Polym. Sci. 1999, 73, 2227−2233.(11) Yao, K. D.; Tu, H. L.; Cheng, F.; Zhang, J. W.; Liu, J. Angew.
Makromol. Chem. 1997, 245, 63−72.(12) Chu, C. H.; Sakiyama, T.; Yano, T. Biosci., Biotechnol., Biochem.
1995, 59, 717−719.(13) Sakiyama, T.; Chu, C. H.; Fujii, T.; Yano, T. J. Appl. Polym. Sci.
1993, 50, 2021−2025.(14) Kumagai, H.; Chu, C. H.; Sakiyama, T.; Ikeda, S.; Nakamura, K.
Biosci., Biotechnol., Biochem. 1996, 60, 1623−1626.(15) Ikeda, S.; Kumagai, H.; Sakiyama, T.; Chu, C. H.; Nakamura, K.
Biosci., Biotechnol., Biochem. 1995, 59, 1422−1427.(16) Chu, C. H.; Kumagai, H.; Sakiyama, T.; Ikeda, S.; Nakamura, K.
Biosci., Biotechnol., Biochem. 1996, 60, 1627−1632.(17) Zezin, A. B.; Kabanov, V. A. Russ. Chem. Rev. 1982, 51, 833−
855.(18) Kabanov, V. A.; Zezin, A. B.; Rogacheva, V. B.; Ryzhikov, S. V.
Dokl. Akad. Nauk SSSR 1982, 267, 862−865.(19) Vasheghani, B. F.; Rajabi, F. H.; Ahmadi, M. H.; Mashhadi, F.
Polym. Bull. 2008, 61, 247−255.(20) Kovacevic, D.; Borkovic, S.; Pozar, J. Colloids Surf., A 2007, 302,
107−112.
Figure 8. FE-SEM images of PMC particles equilibrated at differentsalt concentrations: (a) 1, (b) 5, (c) 100, and (d) 500 mM. Scale bar:500 nm (applies to all images).
200, 118−125.(25) Dautzenberg, H. Macromolecules 1997, 30, 7810−7815.(26) Dragan, S.; Cristea, M. Eur. Polym. J. 2001, 37, 1571−1575.(27) de Vasconcelos, C. L.; Bezerril, P. M.; dos Santos, D. E. S.;
Dantas, T. N. C.; Pereira, M. R.; Fonseca, J. L. C. Biomacromolecules2006, 7, 1245−1252.(28) Bakeev, K. N.; Izumrudov, V. A.; Zezin, A. B.; Kabanov, V. A.
Dokl. Akad. Nauk SSSR 1988, 299, 1405−1408.(29) Pergushov, D. V.; Izumrudov, V. A.; Zezin, A. B.; Kabanov, V.
A. Vysokomol. Soedin., Ser. A Ser. B 1995, 37, 1739−1746.(30) Pergushov, D. V.; Izumrudov, V. A.; Zezin, A. B.; Kabanov, V.
A. Vysokomol. Soedin., Ser. A Ser. B 1993, 35, A844−A849.(31) Izumrudov, V. A.; Kharenko, O. A.; Kharenko, A. V.; Gulyaeva,
Z. G.; Kasaikin, V. A.; Zezin, A. B.; Kabanov, V. A. Vysokomol. Soedin.,Ser. A 1980, 22, 692−699.(32) Nordmeier, E.; Beyer, P. J. Polym. Sci., Part B: Polym. Phys. 1999,
37, 335−348.(33) Wang, H.; Roman, M. Biomacromolecules 2011, 12, 1585−1593.(34) Hejazi, R.; Amiji, M. Chitosan-based delivery system:
Physicochemical properties and pharmaceutical applications. InPolymeric Biomaterials, 2nd ed.; Dumitriu, S., Ed.; Marcel Dekker:New York, 2002; Chapter 10, pp 213−238.(35) Harned, H. S.; Scholes, S. R. Jr. J. Am. Chem. Soc. 1941, 63,
1706−1709.(36) Bratskaya, S. Y.; Avramenko, V. A. Colloid J. 2001, 63, 137−141.(37) Lopez-Leon, T.; Carvalho, E. L. S.; Seijo, B.; Ortega-Vinuesa, J.
L.; Bastos-Gonzalez, D. J. Colloid Interface Sci. 2005, 283, 344−351.
