Monovalent Salt Enhances Colloidal Stability during the Formation of

ARTICLE pubs.acs.org/Langmuir

Monovalent Salt Enhances Colloidal Stability during the Formation of Chitosan/Tripolyphosphate Microgels Yan Huang and Yakov Lapitsky* Department of Chemical and Environmental Engineering, University of Toledo, 3048 Nitschke Hall, 1650 N. Westwood Ave., Toledo, Ohio 43606, United States

bS Supporting Information ABSTRACT: Chitosan micro- and nanoparticles are routinely prepared through ionotropic gelation, where sodium tripolyphosphate (TPP) is added as a cross-linker to dilute chitosan solutions. Despite the wide use of these gel-like particles, their preparation currently relies on trial and error. To address this, we used isothermal titration calorimetry (ITC), dynamic light scattering (DLS), transmission electron microscopy (TEM), and ζ-potential measurements to investigate how the formation, structure, and colloidal stability of chitosan microgels are linked to the molecular interactions that underlie their selfassembly. The strength of the chitosan/TPP interactions was systematically varied through the addition of monovalent salt (NaCl). Remarkably, and contrary to other colloidal systems, this revealed that moderate amounts of NaCl (e.g., 150 mM) enhance the colloidal stability of chitosan/TPP microgels during their formation. This stems from the weakened chitosan/TPP binding, which apparently inhibits the bridging of the newly formed microgels by TPP. The enhanced colloidal stability during the ionic cross-linking process yields microgels with dramatically narrower size distributions, which hitherto have typically required the deacetylation and fractionation of the parent chitosan. Conversely, at high ionic strengths (ca. 500 mM) the chitosan/TPP binding is weakened to the point that the microgels cease to form, thus suggesting the existence of an optimal ionic strength for ionotropic microgel preparation.

1. INTRODUCTION Ionically cross-linked polyelectrolyte micro- and nanoparticles are prepared by titrating bioderived polymers (e.g., chitosan1
4 or alginate5,6) with multivalent ions. These gel-like structures range from tens of nanometers to micrometers in diameter and find countless applications in drug and gene delivery,7,8 foods,9 water disinfection,10 and sorption of toxic metals from aqueous solutions.11 Of these materials, the chitosan/tripolyphosphate (TPP) system is by far the best-characterized. Chitosan/TPP microand nanogels form through the cross-linking of cationic chitosan amines with pentavalent TPP anions.3 Empirically, it is known that colloidally stable chitosan/TPP particles only form when their molecular constituents are mixed within a limited range of stoichiometries3 and that their size and polydispersity depend on chitosan concentration,3,12 degree of deacetylation (DD),1,13 and molecular weight.1,13,14 Likewise, the particle size distribution depends strongly on the chitosan/TPP mixing procedure2,15 and stoichiometry.2,3 The micro- and nanogel size is also sensitive to the ionic strength and pH that the micro- and nanogels are exposed to,2,4 where the variations in size can either reflect changes in the number of polymer chains within each particle or the extent of swelling.4 The micro- and nanogels swell at pH values where they bear a high net charge, collapse when their net charge is reduced, and aggregate into macroscopic flocs when the pH is near their isoelectric point.2,4,16 r 2011 American Chemical Society

Despite this abundance of empirical data, ionotropic microand nanogel preparation relies on trial and error, and the link between their molecular interactions, formation process, and structure remains poorly understood. This is particularly problematic given the batch-to-batch variability of bioderived polymers (such as chitosan or alginate) whose molecular weight and residue distribution varies with the polymer source and preparation conditions. This variability confounds the control over particle size distributions, often leading to high polydispersities12,17 and inconsistent results.1,2,4 To address this, Zhang et al. have recently shown that nanoparticles with comparatively low polydispersities (PDI = 0.17, rather than the typical 0.30
0.4012,17) can be prepared through the deacetylation and fractionation of the parent chitosan molecules.1 This pretreatment yields chitosan molecules that are more uniform in linear charge density and molecular weight and thus results in bettercontrolled particle formation.1 Moreover, the success of this method suggests that the ultimate size distributions of ionically cross-linked particles can be tuned by modulating the molecular interactions of the parent polyelectrolyte molecules during their self-assembly. Received: March 31, 2011 Revised: July 10, 2011 Published: July 12, 2011 10392

dx.doi.org/10.1021/la201194a | Langmuir 2011, 27, 10392–10399

Langmuir

ARTICLE

Figure 1. Representative ITC data for the titration of TPP into chitosan: (a) heat supplied to the sample cell as a function of time and (b) heat absorbed by the TPP/chitosan binding per mole of added TPP as a function of TPP:glucosamine ratio. The horizontal line through the raw data indicates the baseline, while the curve through the integrated data is a guide to the eye.

Because polyelectrolytes are acutely sensitive to their external environments (e.g., pH,18 ionic strength,19 or oppositely charged analytes20,21), their molecular interactions with cross-linking ions can be tuned by modifying the buffer in which the particles form.4 This approach offers two key advantages over modifying the molecular structure of the parent polyelectrolyte molecules: (1) it simplifies the particle preparation procedure by reducing the need for polymer pretreatment (e.g., the deacetylation/fractionation procedure of Zhang et al.1), and (2) the modulation of pH or ionic strength may provide a finer and more-versatile control over the polymer/cross-linker interaction strengths than the synthetic modification method. To this end, we hypothesized that control over the formation and structure of ionically cross-linked polyelectrolyte micro- and nanogels can be achieved through the addition of monovalent salt. Monovalent salt can profoundly weaken the polyelectrolyte/ cross-linker binding strength.21,22 This stems from the competitive binding of multivalent and monovalent ions,21 where the monovalent ions displace their multivalent counterparts from the polyelectrolyte binding sites and cause ionic complexes to dissolve.4,22,23 Herein, we exploited these monovalent salt effects to control the formation and size distributions of ionically cross-linked chitosan/TPP microgels. The effect of monovalent salt (NaCl) on chitosan/TPP binding was studied by isothermal titration calorimetry (ITC) and related to the evolution in the microgel structure, colloidal stability, and surface charge using dynamic light scattering (DLS), transmission electron microscopy (TEM), and ζ-potential measurements. This revealed that the addition of moderate amounts of monovalent salt enhances the colloidal stability during microgel formation and leads to much narrower particle size distributions which hitherto have typically required the deacetylation and fractionation of the parent chitosan molecules.1

2. METHODS 2.1. Materials. All experiments were performed using Millipore Direct-Q 3 deionized water (18.0
18.2 MΩ m). Chitosan (nominal molecular weight of 50
190 kDa), sodium tripolyphosphate (TPP), sodium chloride (NaCl) and acetic acid were purchased from SigmaAldrich (St. Louis, MO). The degree of chitosan deacetylation was estimated at 90% by pH titration, as described previously.24 All materials were used as received without further purification.

2.2. Isothermal Titration Calorimetry. The effect of monovalent salt on the molecular interactions between chitosan and TPP was probed by isothermal titration calorimetry (ITC), using a MicroCal VPITC instrument (GE Healthcare). In each measurement, fifty 5-μL injections of 0.4 wt % TPP (11 mM) were added to a sample cell filled with 1.48 mL of 0.1 wt % chitosan (5.5 mM in its charge-bearing glucosamine monomer units) solution. To maintain a constant pH during the titration (pH 3.8, at which nearly all glucosamine residues are ionized), both solutions were charged with 0.175 wt % acetic acid. The calorimeter measured the heat supplied to the sample cell to keep the cell at a constant temperature. The representative raw data from these experiments is shown in Figure 1a, where each peak reflects a single injection. The integration of each peak over time yielded the heat evolved (or absorbed) per mole of injected TPP as a function of TPP: glucosamine molar ratio within the sample cell (where the glucosamine concentration indicates the concentration of chitosan charge; see Figure 1b). The binding heat was then obtained from this thermogram by subtracting the heat of dilution, which in the case of chitosan/TPP binding was negligibly small. 2.3. Microgel Preparation. Chitosan/TPP microgels were prepared by slow (ca. 20 μL/s) dropwise addition of 0.4 wt % TPP solution to 10 mL of 0.1 wt % chitosan solution, which was mixed with a cylindrical (12 mm  4 mm) magnetic stir bar at 800 rpm inside a 20 mL scintillation vial. To maintain a constant pH of 3.8 during microgel formation, both parent solutions were charged with 0.175 wt % acetic acid. Additionally, to tune the ionic strength, various amounts of NaCl (0
500 mM) were added in matching concentrations to the chitosan and TPP solutions, prior to the dropwise addition process. Once the TPP was added, the mixture was allowed to equilibrate for 10 min before performing the DLS and ζ-potential measurements. 2.4. Dynamic Light Scattering and ζ-Potential Measurements. Microgel size distributions and ζ-potentials were measured

using a Zetasizer Nano ZS (Malvern, UK) dynamic light scattering (DLS) and ζ-potential instrument. The z-average diameters and polydispersity indices (PDIs) were estimated from the autocorrelation functions (not shown) using the cumulant analysis.25 Furthermore, to probe highly polydisperse microgel populations (such as were encountered in the NaCl-free samples), microgel size distributions were also estimated using the multiple narrow modes algorithm.26 The ζ-potentials were then estimated from the electrophoretic mobility measurements using the Smoluchowski model.27 To track the evolution in microgel size, polydispersity, and ζ-potential during their formation process, DLS and ζ-potential measurements were made 10 min after each 100-μL TPP addition. Each measurement was repeated at least three times, thus yielding microgel diameter and ζ-potential data for a 10393

dx.doi.org/10.1021/la201194a |Langmuir 2011, 27, 10392–10399

Langmuir

Figure 2. ITC data for the titration of 0.4 wt % TPP into 0.1 wt % chitosan solution at (red squares) 0 mM NaCl, (blue circles) 30 mM NaCl, and (green triangles) 150 mM NaCl. The solid curves are guides to the eye. range of different TPP:glucosamine molar ratios, with and without added NaCl. 2.5. Transmission Electron Microscopy. To confirm the size distibutions measured by DLS, the microgels were imaged using a Hitachi HD-2300 scanning transmission electron microscope (STEM). One drop of microgel dispersion was placed on a carbon grid, air-dried overnight, and imaged using a 200 kV acceleration voltage. To prevent salt crystal formation on the carbon grid, all microgel dispersions were dialyzed against an excess of NaCl-free acetic acid solution (pH 3.8) prior to being imaged. The dialysis was performed for 2.5 h using SlideA-Lyzer dialysis cassettes (Themo Scientific) with a 2000 MW cutoff. Because the acetic acid evaporated upon drying, this pretreatment yielded TEM samples with minimal salt crystal formation.

3. RESULTS AND DISCUSSION 3.1. Calorimetric Analysis of Chitosan/TPP Binding. The effect of salt on the strength of chitosan/TPP binding was investigated by ITC. Figure 2 shows the enthalpic signatures of chitosan/TPP binding at various NaCl concentrations, where the shape of the curves roughly reflects the first derivative of the binding isotherms (when plotted versus the TPP:glucosamine molar ratio).28 For noncooperative binding with one type of binding site, the enthalpic signal below the saturation point (∼ 0.2:1 TPP:glucosamine molar ratio) is expected to be either constant (at a level equal to the molar enthalpy of chitosan/TPP binding) or to decrease slightly.29 Thus, the increase in the exothermic binding signal in the first half of the ITC curve indicates that the binding strength increases with the TPP coverage (i.e., the binding is cooperative). This is consistent with previous work on the ionic cross-linking of alginate with calcium, where the ITC curves showed similar cooperative features.30 At higher TPP:glucosamine ratios, where the TPP and glucosamine chitosan monomers are mixed at ratios near 0.2:1, there is a dramatic reduction in the exothermic signal. This indicates that the chitosan binding sites become saturated and that each TPP ion occupies approximately five cationic binding sites on the chitosan. Additionally, there is a small endothermic peak at TPP: glucosamine ratios beyond the saturation point. Visual observations at lower ionic strengths (i.e., 0 and 30 mM of added NaCl) suggest that this peak corresponds to the macroscopic precipitation of chitosan/TPP complexes. This agrees well with previous work on the condensation of DNA by multivalent cations, where the precipitation of DNA/cation complexes was accompanied by a similar endothermic peak.21

ARTICLE

Figure 3. Raw ITC data from the 15th TPP injection at (1) 0 mM NaCl, (2) 30 mM NaCl, and (3) 150 mM NaCl. Each curve is normalized to the maximum peak height.

The magnitude of the exothermic binding signal decreased with increasing NaCl concentration. Likewise, the dramatic reduction in the exothermic signal at the saturation point became less abrupt at higher salt concentrations. Both of these effects indicate weaker chitosan/TPP binding28 with increasing NaCl concentration and (especially in the case of the less-abrupt transition at the saturation point) suggest that the addition of monovalent salt inhibits the saturation of chitosan binding sites. This reduction in binding strength is consistent with the previous work on the binding of multivalent cations to DNA and reflects the competitive binding between the monovalent and multivalent ions.21 Furthermore, the small endothermic “precipitation peak” beyond the saturation point diminished with increasing salt concentration and disappeared completely when 150 mM NaCl was used. The disappearance of the precipitation peak agrees well with the visual observations and DLS measurements (see Section 3.2), which showed that precipitation was inhibited even well past the saturation point. This suggests that the weakening of the chitosan/TPP binding by monovalent salt may stabilize the chitosan/TPP microgels against aggregation and precipitation. The analysis of raw ITC data has revealed that NaCl also affects the kinetics of chitosan/TPP binding. This is illustrated in Figure 3, which shows the enthalpic peaks in the raw data that were collected at the three different NaCl concentrations from the 15th TPP injection (where the TPP:glucosamine molar ratio was 0.1:1). To facilitate the comparison between the binding kinetics, the peak magnitudes were normalized to the maximum exothermic signal at each NaCl concentration. At higher NaCl concentrations, significantly longer times were required for the enthalpic signal to return to the baseline. This shows that in the presence of monovalent salt the chitosan/TPP binding is not only weaker but is also slower. By reducing the binding rate relative to the rate at which the chitosan and TPP solutions are mixed, these slower binding kinetics likely lead to a more uniform distribution of chitosan-bound TPP. Thus, by limiting the local saturation of chitosan binding sites—which may occur when the local TPP:glucosamine ratio in the vicinity of the added TPP drop rises above 0.2:1—the slower binding kinetics may enhance the colloidal stability of microgels during their formation. 3.2. Formation and Characterization of Chitosan/TPP Microgels. When TPP was added, the clear chitosan solution became translucent, thus indicating microgel formation. To investigate the microgel formation process, the evolution in microgel size during TPP titration was studied (both in the 10394

dx.doi.org/10.1021/la201194a |Langmuir 2011, 27, 10392–10399

Langmuir

ARTICLE

Figure 4. DLS data showing the evolution in (a) z-average microgel diameter, (b) polydispersity index, and (c) intensity-weighed microgel size distributions (estimated via multiple narrow modes algorithm) for microgels prepared in the presence of (red squares) 0 mM and (blue triangles) 150 mM NaCl. The insets illustrate the colloidal stability of each microgel type beyond the saturation point (at the 0.24:1 TPP:glucosamine ratio). The curves in plots a and b are guides to the eye, and the error bars are standard deviations. The dashed vertical line in plot a denotes the saturation point approximated by ITC.

absence and presence of NaCl) by DLS. The apparent microgel diameter in the NaCl-free solution decreased from approximately 370 to 100 nm when the TPP:glucosamine molar ratio was increased from 0.04 to 0.20:1 (Figure 4a). This is consistent with previous reports, where the microgel size decreased with increasing TPP:chitosan ratio,2 and may be ascribed to the increased ionic cross-link density within the microgel and reduction in microgel charge (both of which diminish swelling31
33). Likewise, the microgel PDI gradually decreased from 0.49 to 0.37 (Figure 4b), which might reflect a progressively more uniform distribution of TPP between the chitosan chains. These high PDI values were confirmed by TEM imaging (performed at a TPP: glucosamine molar ratio of 0.16:1; see Figure 5a), which revealed highly polydisperse microgels ranging from ∼50 to 600 nm in diameter. Moreover, once the binding site saturation point (0.2:1 TPP:glucosamine ratio, as determined by ITC) was exceeded, the microgels began to aggregate and precipitate. This led to a dramatic increase in both size and polydispersity at TPP: glucosamine ratios above 0.2:1. Conversely, microgels formed in 150 mM NaCl solutions remained stably dispersed even when TPP was in excess. Their diameters increased monotonically (from 120 to 170 nm) over the entire range of TPP:glucosamine ratios (from 0.04 to 0.32:1). This trend is opposite to the shrinkage of the particles prepared without added salt and is probably caused by the strongly diminished swelling at higher ionic strengths.31,32,34 Because salt diminishes the osmotic pressure difference between the microgel and supernatant,34 the microgels are already “collapsed” at low TPP:glucosamine ratios, and the addition of TPP has little

impact on their swelling. This view is supported by the greater light scattering intensity exhibited by the microgels in the presence of added salt (see Supporting Information, Figure S1), which suggests a greater refractive index contrast between the microgel and solvent35 and thus a denser microgel structure.32 The slow increase in microgel size in the 150 mM NaCl solution likely reflects an increase in the molecular aggregation number, which in the absence of NaCl was outweighed by the dramatic reduction in swelling with the addition of TPP. The PDIs in the presence of added salt, however, decreased from 0.50 to 0.20 as the TPP:glucosamine ratio changed from 0.04 to 0.16:1 and then remained constant at approximately 0.2 up to a TPP:glucosamine ratio of 0.32:1. The very high PDI values at low TPP:glucosamine molar ratios (which also occurred in the saltfree samples) likely reflect an insufficient amount of TPP to form well-defined microgels. The PDIs at higher TPP:glucosamine molar ratios, however, are dramatically lower than those obtained in the absence of NaCl, and close to that obtained by Zhang et al. through the deacetylation and fractionation of chitosan.1 They are also consistent with the TEM data (obtained at the 0.16:1 TPP:glucosamine ratio; Figure 5b) where—unlike the highly polydisperse microgels prepared without NaCl—all microgels formed in 150 mM NaCl, solutions were 50
250 nm in diameter. This size distribution is consistent with the 215 nm z-average diameter (PDI = 0.21) measured by DLS prior to imaging, where the microgels swelled (and became slightly more polydisperse) during their dialysis due to the removal of salt. Additionally, to confirm that the reduction in polydispersity was not specific to NaCl, microgels were also prepared using 150 mM 10395

dx.doi.org/10.1021/la201194a |Langmuir 2011, 27, 10392–10399

Langmuir

ARTICLE

Figure 5. TEM images of chitosan/TPP microgels prepared using (a) 0 mM and (b) 150 mM NaCl (TPP:glucosamine ratio = 0.16:1).

KCl and yielded the same size distributions as those formed with NaCl (data not shown). This suggests the use of monovalent salt to be an attractive and simple alternative for preparing microgels with narrow size distributions. The lower polydispersities were also evident in the particle size distributions estimated by the multiple narrow modes algorithm (Figure 4c),26 where the size distributions in the presence of NaCl were consistently narrower. Furthermore, without added NaCl the size distributions at TPP:glucosamine ratios of 0.2 and 0.24:1 yielded aggregate peaks in the 1
10 μm range, which indicated macroscopic aggregation. No such aggregates formed, however, when the microgels were prepared in 150 mM NaCl solution. These trends agree with the visual observations (see Figure 4a inset): when the TPP:glucosamine ratio exceeded the saturation point (0.2:1 TPP:glucosamine molar ratio) in the absence of salt, the microgels immediately coagulated and precipitated, while in 150 mM NaCl the microgels remained dispersed (for at least 1 week) up to a TPP:glucosamine ratio of 0.32:1. Although their hydrodynamic diameter and PDI increased over the time scale of days (data not shown), these microgels remained colloidally stable over the shorter minute/hour time scales of their formation (see Figure 4a,b). Importantly, because the TPP:glucosamine ratios that are used for microgel preparation are typically below the binding site saturation point (such that the microgel size distributions remain stable even after days of aging4), microgel polydispersity is controlled by their short-term colloidal stability rather than their long-term colloidal stability. When drops of concentrated TPP are titrated into chitosan solutions, the local TPP:glucosamine ratios around the added drops exceed saturation. These unfavorable compositions persist over short times (seconds or less) until the drops are uniformly mixed with the receiving chitosan solutions. In the absence of NaCl, the local excess of TPP leads to rapid coagulation, which occurs before the TPP can be uniformly mixed with the chitosan. Conversely, in the presence of 150 mM NaCl, the rate of mixing exceeds the slow rate of coagulation and keeps the newly formed microgels stably dispersed. This salt-mediated reduction in coagulation rate is consistent with the binding kinetics measured by ITC (Figure 3), where the rates of TPP binding were diminished by the added NaCl. Thus, by limiting microgel coagulation, the use of monovalent salt during ionotropic gelation yields microgels with narrow size distributions.

Besides reducing coagulation and polydispersity, the addition of NaCl makes the microgel size distributions more reproducible. This is evident from the much smaller standard deviations in the z-average microgel diameter (given by the error bars in Figure 4a). The poor reproducibility in the case of NaCl-free solutions likely stems from the coagulation of microgels into larger particles, which is difficult to control because it is very sensitive to the mixing process. In the presence of 150 mM NaCl, however, coagulation is prevented and the z-average microgel diameter varies little between microgel batches. The differences in the polydispersity and colloidal stability likely stem from the weaker chitosan/TPP binding, which may prevent microgel coagulation in two ways: (1) it may enhance electrostatic stabilization by inhibiting charge neutralization, and (2) it may prevent bridging flocculation25 of chitosan microgels by the TPP. When droplets of TPP solution are added to the chitosan solution, local binding site saturation may be reached around the added drops (even when the overall TPP:glucosamine ratio is very low). Consequently, the neutralization of the surface charge on the newly formed microgels may undermine their colloidal stability by diminishing electrostatic repulsion (thereby leading to coagulation and high polydispersities).4,25 The other possibility is that the free TPP ions might form crosslinks between two microgels (rather than two polymer chains within the same microgel), resulting in bridging flocculation. In this case, the salt-mediated weakening of the chitosan/TPP binding may inhibit interparticle cross-linking. Interestingly, when the NaCl concentration was increased further to 500 mM, the microgels ceased to form. This likely reflects the further weakening of chitosan/TPP binding and suggests that at high monovalent salt concentrations the polymer/cross-linker binding is too weak to form stable particles. However, when preformed microgels are placed in concentrated NaCl solutions—with concentrations as high as 1.5 M—the microgels maintain both their colloidal and phase stability (i.e., microgel size and polydispersity remains nearly constant as the ionic strength is gradually increased from 150 mM to 1.5 M; see Supporting Information, Figure S2). The stable microgel dispersions in 1.5 M NaCl indicate that their colloidal stability is insensitive to changes in electrostatic repulsion and suggest bridging flocculation to be their main coagulation mechanism. Moreover, their high phase stability at high NaCl concentrations show that, although 10396

dx.doi.org/10.1021/la201194a |Langmuir 2011, 27, 10392–10399

Langmuir

ARTICLE

Figure 6. Evolution in the microgel (a) ζ-potential and (b) surface charge (estimated by the Gouy
Chapman theory) during the titration of TPP into chitosan: (red square) without NaCl and (blue triangle) with 150 mM NaCl. The curves are guides to the eye and the error bars are standard deviations.

NaCl can inhibit the cross-linking of chitosan by TPP, the cross-links are very stable to NaCl once they form. In summary, the addition of monovalent salt has two effects on microgel preparation: (1) reduction in polydispersity at low-tomoderate ionic strengths and (2) inhibition of microgel formation at high ionic strengths. The former effect provides better control over microgel formation and is thus desirable, whereas the latter effect disrupts microgel formation and is therefore undesirable. This suggests that there is an optimal ionic strength for ionotropic microgel formation. 3.3. Evolution in ζ-Potential. To further probe the origins of colloidal stability in chitosan/TPP microgel dispersions, the evolution in microgel ζ-potentials during the TPP titration was investigated. Due to the cationic charge on the chitosan molecules, the ζ-potential was positive (see Figure 6a) and was diminished by the binding of anionic TPP. In microgels prepared without NaCl, this decrease was dramatic (from 55 to 10 mV as the TPP:glucosamine molar ratio was increased from 0.05 to 0.40:1). Conversely, in the presence of 150 mM NaCl, the ζ-potential decrease was more gradual (from 27 to 15 mV as the TPP:glucosamine molar ratio was increased from 0.05 to 0.40:1). Despite the 2-fold excess of TPP at the end of the titration (binding site saturation occurs at the TPP:glucosamine ratio of 0.20:1; see Section 3.1), the microgels maintained a positive ζ-potential. This suggests that the microgel surface is difficult to neutralize through TPP binding, possibly because the surfacebound TPP is less likely to form stable chitosan cross-links than the TPP in the microgel bulk. Importantly, although most ζ-potential values were lower in 150 mM NaCl than in salt-free solutions, they did not indicate lower surface charge densities. This was shown by the Gouy
Chapman theory, which (although strictly a model for hard interfaces) served as a semiquantitative argument for the soft chitosan microgels. For a fixed surface potential (ψ0, where ψ0 ∼ ζ-potential), it predicts the surface charge density to increase with the square root of salt concentration, n∞, as25 σ ¼ ð2kB Tn∞ εÞ1=2 ½expðzeψ0 =2kB TÞ
expð
zeψ0 =2kB TÞ ð1Þ where kB is the Boltzmann constant, T is the absolute temperature, ε is the solvent dielectric constant, e is the charge of a single proton, and z is the valence of the added salt (equal to 1 for

NaCl). This semiquantitative analysis suggests that, despite the lower ζ-potential, the surface charge density may be higher in the presence of 150 mM NaCl than in the NaCl-free system (see Figure 6b and Supporting Information for calculation details). Additionally, the ζ-potential measurements in Figure 6a suggest that the enhanced colloidal stability of chitosan/TPP microgels in the presence of NaCl cannot be explained through electrostatic repulsion alone. The inset in Figure 4 shows that when TPP and chitosan are mixed at ratios above the binding site saturation point (0.2:1 TPP:glucosamine molar ratio) the microgels remain dispersed in the presence of 150 mM NaCl but coagulate if no NaCl is added. Yet, at the saturation point both the ζ-potential (33 mV versus 20 mV) and Debye length (4.1 nm versus 0.79 nm) in the NaCl-free chitosan/TPP dispersions are higher, indicating that the addition of NaCl weakens the electrostatic repulsion between the microgels. Consequently, if the colloidal stability stemmed from electrostatic repulsion, the addition of NaCl would reduce the colloidal stability rather than enhancing it. This, combined with the stability of these microgel dispersions in 1.5 M NaCl, suggests the salt-mediated inhibition of bridging flocculation to be the more likely cause of the anomalous salt effects on the colloidal stability of chitosan/ TPP microgels. This interpretation is further supported by the insensitivity of the microgel ζ-potential to TPP in the presence of 150 mM NaCl, which suggests that few TPP ions are binding to the microgel surface. 3.4. Effect of Monovalent Salt on Preformed Microgels. To verify that the narrower size distributions in the presence of NaCl were due to enhanced colloidal stability (rather than saltinduced structural rearrangements), the effects of NaCl on the size distributions of preformed microgels were investigated. Microgels were prepared at the TPP:glucosamine molar ratio of 0.10:1 in 150 mM NaCl solution and were diluted in deionized water to lower (10
100 mM) NaCl concentrations. After 1 h of equilibration, their size distributions were measured by DLS. Similarly, different amounts of NaCl (25
150 mM) were added to microgels prepared in NaCl-free solutions and also sized by DLS after 1 h of equilibration. Figure 7 shows that the sizes (140
160 nm) and low PDI values (0.19
0.25) of microgels prepared in 150 mM NaCl solutions changed little when the ionic strength was decreased and exhibited a subtle minimum in both size and polydispersity at approximately 50 mM NaCl. Conversely, when NaCl was added to microgels prepared 10397

dx.doi.org/10.1021/la201194a |Langmuir 2011, 27, 10392–10399

Langmuir

ARTICLE

Figure 7. The effect of NaCl on the (a) z-average diameter and (b) polydispersity index of microgels prepared in (red square) salt-free solutions and (blue triangle) 150 mM NaCl. The curves are guides to the eye and the error bars are standard deviations.

without NaCl, the microgel size and polydispersity decreased. Their PDIs, however (which ranged between 0.31 and 0.37), were still much higher than those of microgels prepared using NaCl. To ensure that the microgel size distributions did not evolve over time, the microgels were characterized again after 2 days and showed no change in z-average diameter and PDI. If the narrower size distributions were caused only by saltinduced structural rearrangements, microgels prepared in the absence of NaCl were expected to have similar size distributions to those prepared in the presence of NaCl when placed in solutions of matching ionic strength. Conversely, if the narrower size distributions were caused only by the enhanced colloidal stability, the changes in NaCl concentrations were expected to have little impact on microgel PDIs. The reduction in the PDI upon the addition of NaCl to particles prepared in salt-free media suggests that structural rearrangement plays a partial role in the reduction of polydispersity. This might reflect the inhomogeneous cross-link density between the microgels, which leads to a mixture of highly swollen larger particles and less-swollen smaller particles; because the microgels collapse in the presence of NaCl (see Figure 7a), the dramatic collapse of the loosely cross-linked larger microgels (and subtle shrinkage of densely cross-linked smaller microgels) likely leads to a reduction in PDI. This view is further supported by the structural transitions of microgels formed in the presence of NaCl (Figure 7), where both the z-average diameter and PDI increase slightly in the limit of low NaCl concentration (presumably due to the reswelling of microgels with looser cross-link densities). Despite these structural rearrangements, however, the PDIs of microgels formed in 150 mM NaCl were much lower than those formed in NaCl-free solution at all salt concentrations. This indicates the low PDIs achieved in the presence of NaCl to be predominantly a kinetic effect, achieved by enhancing colloidal stability during microgel formation.

4. CONCLUSIONS Chitosan/TPP microgels with narrow size distributions were prepared through ionotropic gelation. This was achieved by amplifying their colloidal stability which, unlike in most other systems, is enhanced by monovalent salt. The enhanced colloidal stability stems from the weakened chitosan/TPP binding, which appears to inhibit the bridging flocculation of microgels by TPP. By preventing microgel coagulation during their formation, the use of monovalent salt results in microgels that are much less polydisperse than those prepared under salt-free conditions.

Furthermore, unlike in salt-free systems, where the microgel dispersions are only stable when the TPP:glucosamine ratio is below the saturation point, NaCl allows the dispersions to remain stable even when TPP is in excess. In the limit of very high NaCl concentrations, however, chitosan/TPP binding is weakened to the point that microgels cease to form, suggesting that there is an optimal ionic strength for ionotropic microgel preparation. These findings demonstrate that the formation and structure of ionotropically prepared microgels can be easily tuned by modulating the polyelectrolyte/cross-linker interaction strength.

’ ASSOCIATED CONTENT

bS

Supporting Information. Light scattering intensity from chitosan/TPP microgels formed in 0 and 150 mM NaCl solutions, DLS measurements of microgel diameters and PDIs at high (150
1500 mM) NaCl concentrations, and details on the microgel surface charge density calculations. This material is available free of charge via the Internet at http://pubs.acs.org.

’ AUTHOR INFORMATION Corresponding Author

*E-mail: [email protected].

’ ACKNOWLEDGMENT We are grateful to the University of Toledo’s College of Engineering, for supporting this work, and to Dr. Joseph Lawrence, for assistance with TEM imaging. ’ REFERENCES (1) Zhang, H.; Oh, M.; Allen, C.; Kumacheva, E. Monodisperse chitosan nanoparticles for mucosal drug delivery. Biomacromolecules 2004, 5, 2461–2468. (2) Gan, Q.; Wang, T.; Cochrane, C.; McCarron, P. Modulation of surface charge, particle size and morphological properties of chitosan
TPP nanoparticles intended for gene delivery. Colloids Surf., B 2005, 44, 65–73. (3) Calvo, P.; Remunan-Lopez, C.; Vila-Jato, J. L.; Alonso, M. J. Novel hydrophilic chitosan
polyethylene oxide nanoparticles as protein carriers. J. Appl. Polym. Sci. 1997, 63, 125–132. (4) Lopez-Leon, T.; Carvalho, E. L. S.; Seijo, B.; Ortega-Vinuesa, J. L.; Bastos-Gonzalez, D. Physicochemical characterization of chitosan nanoparticles: Electrokinetic and stability behavior. J. Colloid Interface Sci. 2005, 283, 344–351. 10398

dx.doi.org/10.1021/la201194a |Langmuir 2011, 27, 10392–10399

Langmuir (5) Rajaonarivony, M.; Vauthier, C.; Couarraze, G.; Puisieux, F.; Couvreur, P. Development of a new drug carrier made from alginate. J. Pharm. Sci. 1993, 82, 912–17. (6) De, S.; Robinson, D. Polymer relationships during preparation of chitosan
alginate and poly-L-lysine
alginate nanospheres. J. Controlled Release 2003, 89, 101–112. (7) Janes, K. A.; Calvo, P.; Alonso, M. J. Polysaccharide colloidal particles as delivery systems for macromolecules. Adv. Drug Delivery Rev. 2001, 47, 83–97. (8) Wagner, E.; Kloeckner, J. Gene delivery using polymer therapeutics. Adv. Polym. Sci. 2006, 192, 135–173. (9) Jang, K.-I.; Lee, H. G. Stability of chitosan nanoparticles for L-ascorbic acid during heat treatment in aqueous solution. J. Agric. Food Chem. 2008, 56, 1936–1941. (10) Li, Q.; Mahendra, S.; Lyon, D. Y.; Brunet, L.; Liga, M. V.; Li, D.; Alvarez, P. J. J. Antimicrobial nanomaterials for water disinfection and microbial control: Potential applications and implications. Water Res. 2008, 42, 4591–4602. (11) Qi, L. F.; Xu, Z. R. Lead sorption from aqueous solution on chitosan nanoparticles. Colloid Surf. A 2004, 251, 183–190. (12) Tsai, M. L.; Bai, S. W.; Chen, R. H. Cavitation effects versus stretch effects resulted in different size and polydispersity of ionotropic gelation chitosan
sodium tripolyphosphate nanoparticle. Carbohydr. Polym. 2008, 71, 448–457. (13) Huang, M.; Khor, E.; Lim, L.-Y. Uptake and cytotoxicity of chitosan molecules and nanoparticles: Effects of molecular weight and degree of deacetylation. Pharm. Res. 2004, 21, 344–353. (14) Janes, K. A.; Alonso, M. J. Depolymerized chitosan nanoparticles for protein delivery: Preparation and characterization. J. Appl. Polym. Sci. 2003, 88, 2769–2776. (15) Nasti, A.; Zaki, N. M.; de, L. P.; Ungphaiboon, S.; Sansongsak, P.; Rimoli, M. G.; Tirelli, N. Chitosan/TPP and chitosan/TPP
hyaluronic acid nanoparticles: Systematic optimization of the preparative process and preliminary biological evaluation. Pharm. Res. 2009, 26, 1918–1930. (16) Zhang, H.; Mardyani, S.; Chan, W. C. W.; Kumacheva, E. Design of biocompatible chitosan microgels for targeted pH-mediated intracellular release of cancer therapeutics. Biomacromolecules 2006, 7, 1568–1572. (17) Dudhani, A. R.; Kosaraju, S. L. Bioadhesive chitosan nanoparticles: Preparation and characterization. Carbohydr. Polym. 2010, 81, 243–251. (18) Tan, B. H.; Tam, K. C. Review on the dynamics and microstructure of pH-responsive nano-colloidal systems. Adv. Colloid Interface Sci. 2008, 136, 25–44. (19) Kohler, K.; Biesheuvel, P. M.; Weinkamer, R.; Mohwald, H.; Sukhorukov, G. B. Salt-induced swelling-to-shrinking transition in polyelectrolyte multilayer capsules. Phys. Rev. Lett. 2006, 97, 188301/1–188301/4. (20) Hansson, P. Interaction between polyelectrolyte gels and surfactants of opposite charge. Curr. Opin. Colloid Interface Sci. 2006, 11, 351–362. (21) Matulis, D.; Rouzina, I.; Bloomfield, V. A. Thermodynamics of DNA binding and condensation: Isothermal titration calorimetry and electrostatic mechanism. J. Mol. Biol. 2000, 296, 1053–1063. (22) Dautzenberg, H.; Kriz, J. Response of polyelectrolyte complexes to subsequent addition of salts with different cations. Langmuir 2003, 19, 5204–5211. (23) Dautzenberg, H.; Rother, G. Response of polyelectrolyte complexes to subsequent addition of sodium chloride: Time-dependent static light scattering studies. Macromol. Chem. Phys. 2004, 205, 114–121. (24) Lapitsky, Y.; Zahir, T.; Shoichet, M. S. Modular biodegradable biomaterials from surfactant and polyelectrolyte mixtures. Biomacromolecules 2008, 9, 166–74. (25) Hiemenz, P. C.; Rajagopalan, R. Principles of Colloid and Surface Chemistry, 3rd ed.; Marcel Dekker, Inc.: New York, 1977. (26) Nobbmann, U.; Morfesis, A., Characterization of nanoparticles by light scattering. Mater. Res. Soc. Symp. Proc. 2008, 1074E, Paper # 1074-I10
45.

ARTICLE

(27) Hessemann, R. The zeta potential. A measuring quantity for determination of the disperse system stability. Coating 2002, 35, 284–286. (28) O’Brien, R.; Haq, I. In Applications of Biocalorimetry: Binding, Stability and Enzyme Kinetics; John Wiley & Sons Ltd.: New York, 2004; pp 3
34. (29) Velazquez-Campoy, A. Ligand binding to one-dimensional lattice-like macromolecules: Analysis of the McGhee
von Hippel theory implemented in isothermal titration calorimetry. Anal. Biochem. 2006, 348, 94–104. (30) Fang, Y.; Al-Assaf, S.; Phillips, G. O.; Nishinari, K.; Funami, T.; Williams, P. A.; Li, L. Multiple steps and critical behaviors of the binding of calcium to alginate. J. Phys. Chem. B 2007, 111, 2456–2462. (31) Fujii, S.; Dupin, D.; Araki, T.; Armes, S. P.; Ade, H. First direct imaging of electrolyte-induced deswelling behavior of pH-responsive microgels in aqueous media using scanning transmission X-ray microscopy. Langmuir 2009, 25, 2588–2592. (32) Ogawa, K.; Sato, S.; Kokufuta, E. Formation of intra- and interparticle polyelectrolyte complexes between cationic nanogel and strong polyanion. Langmuir 2005, 21, 4830–4836. (33) Flory, P. J.; Rehner, J., Jr. Statistical mechanics of cross-linked polymer networks. II. Swelling. J. Chem. Phys. 1943, 11, 521–6. (34) Capriles-Gonzalez, D.; Sierra-Martin, B.; Fernandez-Nieves, A.; Fernandez-Barbero, A. Coupled deswelling of multiresponse microgels. J. Phys. Chem. B 2008, 112, 12195–12200. (35) Hoare, T.; Pelton, R. Functionalized microgel swelling: Comparing theory and experiment. J. Phys. Chem. B 2007, 111, 11895–11906.

10399

dx.doi.org/10.1021/la201194a |Langmuir 2011, 27, 10392–10399