Article pubs.acs.org/Langmuir

Partitioning of Humic Acids between Aqueous Solution and Hydrogel. 2. Impact of Physicochemical Conditions Katarzyna Zielińska,†,‡ Raewyn M. Town,*,§ Kamuran Yasadi,† and Herman P. van Leeuwen† †

Laboratory of Physical Chemistry and Colloid Science, Wageningen University, Dreijenplein 6, 6703 HB Wageningen, The Netherlands ‡ School of Biological and Chemical Sciences, Queen Mary University of London, Mile End Road, London E1 4NS, United Kingdom § Department of Physics, Chemistry and Pharmacy, University of Southern Denmark, Campusvej 55, DK-5230 Odense, Denmark S Supporting Information *

ABSTRACT: The effects of the physicochemical features of aqueous medium on the mode of partitioning of humic acids (HAs) into a model biomimetic gel (alginate) and a synthetic polyacrylamide gel (PAAm) were explored. Experiments were performed under conditions of different pH and ionic strength as well as in the presence or absence of complexing divalent metal ions. The amount of HA penetrating the gel phase was determined by measuring its natural fluorescence by confocal laser scanning microscopy. In both gel types, the accumulation of HA was spatially heterogeneous, with a much higher concentration located within a thin film at the gel surface. The thickness of the surface film (ca. 15 μm) was similar for both types of gel and practically independent of pH, ionic strength, and the presence of complexing divalent metal ions. The extent of HA accumulation was found to be dependent on the composition of the medium and on the type of gel. Significantly more HA was accumulated in PAAm gel as compared to that in alginate gel. In general, more HA was accumulated at lower background salt concentration levels. The distribution of different types of HA species in the gel body was linked to their behavior in the medium and the differences in physicochemical conditions inside the two phases.

■

INTRODUCTION Humic substances are naturally occurring organic materials, nominally divided into humic acids (HAs, which precipitate at pH 1) and fulvic acids (FAs, which are soluble at all pH values), that constitute a major pool of organic carbon in terrestrial and aquatic environments. They feature a heterogeneous, colloidal, polydisperse, and polyelectrolytic character and a mixed aliphatic and aromatic nature,1,2 enabling them to interact with hydrophilic and hydrophobic compounds and regulate the bioavailability thereof.3,4 Humic substances contain many ionizable functional groups, including primarily carboxylic acids and phenols as well as amines, thiols, and, to a lesser extent, sulfonic acids. Furthermore, they are very dynamic in nature: humic substances respond to changes in environmental conditions by physicochemical reorganization in order to adopt the most energetically favorable conformation under the given pH and ionic strength. Much work has been done to understand the processes of complexation, aggregation, and sorption of humic substances. Recently, this was complemented with research on the penetration of soil HAs into gel phases,5 namely, a naturally occurring biogel, alginate, and a synthetic polyacrylamide gel, PAAm. Biogels are ubiquitous in the aquatic environment, e.g., in the form of biofilm matrices6 and as components of cell walls. Like HA, they play an important role in regulating the © 2014 American Chemical Society

speciation of trace compounds in aquatic systems and biouptake of nutrients and pollutant species. Characterization of the interactions between HA and biogel components, and the fate of associated compounds, is fundamental to quantifying links between chemical reactivity and bioreactivity of target compounds. Alginate is a useful model biogel: it is a major component of cell walls of brown algae and displays a remarkably high affinity for divalent metal cations.7,8 Alginate contains a mixture of 1,4linked β-D-mannuronic and α-L-guluronic acid residues and exhibits a pH-dependent negative electric charge. This results in a Donnan potential difference between an alginate gel and its aqueous surroundings. The magnitude of the Donnan potential is dependent on both pH and ionic strength.9,10 As a consequence of Donnan partitioning, the concentrations of cations in the gel phase are enhanced, and those of anions are reduced, relative to those in the surrounding aqueous medium. Alginate gels are held together by noncovalent bonds that can easily break and reform in response to changing physicochemical conditions.11 Thus, in addition to variations of the Donnan potential, specific structural changes within the gel may occur as Received: November 9, 2014 Revised: December 5, 2014 Published: December 5, 2014 283

dx.doi.org/10.1021/la504393r | Langmuir 2015, 31, 283−291

Langmuir

Article

Figure 1. Impact of ionic strength on the logarithmic distribution of the radius, rp, of soil HA particles in 0.04 kg m−3 dispersion at pH 6 in the absence of Cd(II). For clarity, the curves show the envelope of the original bar diagram output from the DTS nano software. disks per sample volume) containing 100 mL of 0.04 kg m−3 HA dispersion at the given pH and ionic strength. The alginate gels lost their structural integrity in 100 mol m−3 electrolyte; hence, HA accumulation studies with this gel could not be performed at this ionic strength. The gels were freely floating in the sample solution, without forced convection. In view of the following paper in this series on the gel/water partitioning of humic acid complexes of metal ions, some of the samples were given a total Cd(II) concentration of 10−2 mol m−3 by addition of Cd(NO)3. The pertaining metal-to-ligand ratio is kept much lower than unity because of the ensuing simplifications in electrochemical data analysis and interpretation of the ion complexation by the heterogeneous humic acid complexant. The conditions correspond to ca. 50% of the Cd being complexed by HA at pH 6 and ionic strength 10 mol m−3. The penetration of HA into the gels was followed by confocal laser scanning microscopy (CLSM): depthresolved (z-stacked) fluorescence images were taken with steps of 5 or 8 μm within the gel. Dynamic Light Scattering (DLS). DLS measurements of the size of HA aggregates in the aqueous medium were performed with a Zetasizer Nano ZS from Malvern Instruments along with the DTS (Nano) program for data evaluation and transformation of the intensity data to size distributions on the basis of volume and number of particles. Measurements were performed on freshly prepared dispersions, and the size distributions remained constant over several days. Since we consider the concentration of HA in terms of mass, the volume distribution is used for subsequent computations of, e.g., the average diffusion coefficient. The determination of particle size from DLS measurements is somewhat approximate for the case of soft, heterogeneous HA particles. For present purposes, the results are used to identify any trends in particle size as a function of physicochemical conditions. Before measurements, the samples were filtered through a 0.2 μm filter directly attached to the optical cell. The temperature of the scattering cell was controlled at 25.0 ± 0.1 °C. Measurements were performed in triplicate. Confocal Laser Scanning Microscopy (CLSM). The concentration of HA in the gels was measured by CLSM as described previously.5 Briefly, after exposure to HA, the gels were placed on a microscopic slide and immediately analyzed on a Zeiss Axiovert 200M microscope (Carl Zeiss, Germany) equipped with an argon laser set at 488 nm (stabilized for 30 min prior to measurements). Sets of front views and cross-sectional views through the gels were collected. Threedimensional reconstructions from z-stack images were created using Zeiss ZEN software. The concentrations of HA sorbed by the gels were determined on the basis of fluorescent calibration in aqueous dispersion. Additionally, the fluorescence intensity of all HA dispersions used in the gel equilibration studies was measured. Images were taken for a constant scanned area with two different pinhole settings, resulting in an optical window with 5 or 8 μm slice thickness. Calibration curves are given in a previous publication5 and in the Supporting Information, Figure S1. As detailed in previous work, with the given microscope settings, the CLSM response loses its proportionality to the HA concentration at values greater than 0.7 kg m−3.5 This value is denoted as cHA,limit in the figures.

a function of ionic strength, resulting in swelling or shrinking of the gel matrix and corresponding changes in gel density.9 In contrast, hydrogels of cross-linked polyacrylamide, such as PAAm, currently applied in, e.g., dynamic trace metal speciation analysis,12 are considered to be effectively uncharged at ionic strengths greater than ca. 1 mol m−3.13 Furthermore, there is no evidence for swelling/shrinking of PAAm as a function of pH or background electrolyte concentration. There is a paucity of information on the pore size distribution in alginate and PAAm gels and its profiling at the gel/medium interface. Studies on the penetration of various entities into gels have shown, e.g., that latex particles of radii up to 130 nm can enter PAAm,14 whereas bovine serum albumin fully penetrates alginate gel.15 As an extension of our previous studies,5 the present work explores the effects of the solution pH, background electrolyte concentration, and the presence of complexing metal ions on the partitioning of HA species between aqueous medium and two types of gels, i.e., alginate and PAAm.

■

EXPERIMENTAL SECTION

Materials. The humic acid was extracted from Tongbersven forest soil (Oisterwijk, The Netherlands). Its characterization has been reported in the literature.5,16 D-Glucono-δ-lactone, solid CaCO3, and Na-alginate (A-2033, lot 128F0050) were from Sigma-Aldrich. All solutions and gels were prepared with distilled, deionized water from an Easy Pure UV system (Barnstead, resistivity < 18 MΩ cm). Ca(NO3)2 was obtained from J.T. Baker BV, NaNO3, from Fluka, and NaOH, HNO3, and Cd(NO3)2, from Sigma-Aldrich. Stock HA dispersions of 1 kg m−3 were prepared at pH 9.6. Subsequent working solutions (0.04 kg m−3) were prepared by dilution in NaNO3. The final ionic strength, I, of sample solutions was equal to 1, 10, or 100 mol m−3. The pH was adjusted to the desired value with NaOH and HNO3, after which it remained constant throughout the duration of the experiments. Gel Preparation. Alginate gels were prepared according to a method described in the literature.9 Briefly, 1% sodium alginate solutions were gelated in small cylindrical wells, ca. 1 cm in diameter and 2 cm in depth, with CaCO3 (15 mol m−3) and D-glucono-δlactone (30 mol m−3) and were then equilibrated in 50 mol m−3 Ca(NO3)2 + 20 mol m−3 NaNO3. To remove excess calcium (an interferent due to its complex formation with HA), the gels were successively equilibrated with 100 mol m−3 and 0.1 mol m−3 NaNO3 for no longer than 2 days. The polyacrylamide (PAAm) gels were homemade and stored until use, according to the protocol provided by DGT Research (Lancaster, UK).17 They were cast at 0.50 mm thickness, finally resulting in aqueous gel disks with 0.84 mm thickness and a radius of 12 mm. The water volume fraction for this open pore gel type is 0.95,17 and the structural charge density is ca. 1 mol m−3.13 Permeation Measurements. Both gel types, pre-equilibrated in the electrolyte medium, were placed for 1 day in separate vials (2 gel 284

dx.doi.org/10.1021/la504393r | Langmuir 2015, 31, 283−291

Langmuir

■

Article

Table 1. Influence of pH on the Peak Radii, rpeak, of Size Classes of HA Entities in the Presence of Cd2+ Ionsa

RESULTS AND DISCUSSION

It is known that a number of physicochemical properties of HAs may be quite dynamic in nature. In response to varying conditions, they may undergo changes in, e.g., their conformation, aggregation/repeptization, and chemical association/dissociation.16,18−20 In case of the soil HA studied in the present work, this behavior was confirmed by measurements of aggregate size and fluorescence activity via dynamic light scattering (DLS) and confocal laser scanning microscopy (CLSM) as a function of pH and ionic strength. Size Distribution of HA Particles in Aqueous Solution. HA is amphiphilic in nature: in water, it can aggregate spontaneously, forming micelle-like aggregates, so-called pseudomicelles, that are comparable to those formed by surfactants.21−23 The extent to which HAs form such aggregates depends on the solution conditions and the type of HAs. The relatively large soil HAs have been shown to readily form such structures.24 The tendency for HA to form pseudomicelles is generally increased in the presence of complexing cations due to reduction of both the charge density and the hydrophilicity of HA.24,25 Effect of Ionic Strength on the Size Distribution of HA Particles. Figure 1 shows the dependence of the size distribution of HA particles on ionic strength at pH 6. The presence of 10−2 mol m−3 Cd(II) was found to have no systematic effect on the size distribution. At all examined ionic strengths, HA dispersions feature a bimodal size distribution, i.e., with one size class centered at a radius of ca. 20−40 nm and another one at 100−300 nm. The relative importance of the two size categories depends on the concentration of background electrolyte: in 1 mol m−3 electrolyte, HA particle size classes have peak radii, rpeak, of ca. 20 and 170 nm, whereas in 100 mol m−3 electrolyte, these values are ca. 42 and 300 nm, respectively. Increasing the ionic strength apparently enhances the screening of intraparticulate HA charges and thus favors hydrophobic interactions, which generally leads to increased particle aggregation (Figure 1). Other studies have revealed similar dependencies of the size of HA aggregates on electrolyte concentration.19,26 Effect of pH on the Size Distribution of HA Particles. For an ionic strength of 10 mol m−3, the changes in particle radius with varying pH are modest (Table 1). The slight increase in size on going from pH 6 to 8 may be due to the growing charge density that generates stronger intraparticulate electrostatic repulsion, resulting in expansion of the particulate entities with some fragmentation eventually occurring at pH 9, yielding a decrease in particle size. The behavior of the present soil humic acid is similar to the pH-dependent aggregation of peat humic acids, which has been described in terms of hydrophobic interactions.19,27 Fluorescence Intensity of HA. Effect of Ionic Strength on Fluorescence Intensity of HA. Changes in the local environment of HA functional groups will not only influence the size of the HA entities but also their spectroscopic characteristics, i.e., their specific fluorescence coefficient.28,29 The fluorescence intensity of HA dispersions decreases with increasing ionic strength (Figure S2a), with the most significant drop of ca. 50% observed upon increasing the ionic strength from 10 to 100 mol m−3. This observation may be connected to the larger HA aggregates present in the 100 mol m−3 electrolyte medium (Table 1): larger particles may be less fluorescent due to self-quenching arising from the reduced distance between

pH

rpeak ± SD (nm) by volume

6 8

26 ± 3 (85%) 185 38 ± 4 (65%) 193

3 4 6 8 9

23 40 32 37 27

6

59 ± 7 (38%) 296

± ± ± ± ±

2 3 4 4 5

(84%) (64%) (79%) (73%) (80%)

113 218 191 202 146

rpeak ± SD (nm) by number

I = 1 mol m−3 ± 17 (15%) 25 ± 28 (35%) 35 I = 10 mol m−3 ± 18 (15%) 21 ± 32 (36%) 36 ± 20 (21%) 30 ± 24 (27%) 32 ± 42 (20%) 26 I = 100 mol m−3 ± 38 (62%) 67

± 3 (99%) 133 ± 12 (1%) ± 3 (99%) 150 ± 22 (1%) ± ± ± ± ±

2 2 3 5 3

(99%) (99%) (99%) (99%) (99%)

93 ± 23 (1%) 160 ± 15 (1%) 158 ± 20 (1%) 145 ± 23 (1%) 133 ± 37 (1%)

± 7 (99%) 274 ± 24 (1%)

The total concentration of Cd(II), c*Cd(II), is 10−2 mol m−3, and that of * , is 0.04 kg m−3. In each case, the percentage of the size HA, cHA,aq population is shown in brackets. The distributions are represented in terms of pertaining HA volume as well as in terms of the number of individual particles within the given size categories. SD = standard deviation. a

donors and acceptors.30 Indeed, smaller HA entities exhibiting greater fluorescence intensity than larger ones31 may merely be a consequence of the reduced molecular heterogeneity in the separated size fractions, which favors increased fluorescence quantum yield.30,32,33 Effect of Cd(II) on Fluorescence Intensity of HA. For the chosen metal-to-HA concentration ratio, there was no consistent trend in the effect of cadmium ions on the fluorescence intensity of HA (Figure S2a). Although fluorescence quenching has been used to follow the extent of metal ion complexation by HA,34,35 significant effects for Cd(II) are reported only for much greater metal-to-ligand ratios than those used in the present work.36,37 Furthermore, while binding of divalent metal ions may invoke aggregation of HA, this phenomenon is more evident for the strongly bound Cu2+ than for the fairly weakly bound Cd(II).38 We did not observe any significant effect of Cd2+ on the size distribution of HA particles (see above). Effect of pH on Fluorescence Intensity of HA. Figure S2b shows that in the presence of Cd(II) at an ionic strength of 10 mol m−3 there is an overall decrease in the intensity of HA fluorescence on going from pH 7 to 3, with the most significant drop occurring between pH 4 and 3. Over this pH range, there will be a change in the extent to which HA is complexed with Cd(II), but no pronounced change was observed in the size distribution of the HA aggregates (Table 1). Furthermore, as noted in the previous section, Cd(II) had no significant effect on the HA fluorescence intensity under the conditions considered herein. Others have reported that in the absence of complexing metal ions the fluorescence intensity of HA decreases with decreasing pH,29 most probably due to protonation/deprotonation of individual fluorophores, e.g., some may fluoresce only as the conjugate base or conjugate acid, or both forms may fluoresce.39 Accumulation of HA in the Gel Phase: Equilibrium Concentration Profiles. Previous work established that HAs partition into PAAm and alginate gels in a heterogeneous manner, i.e., the average concentration in a thin film, adjacent to the interface with the medium, is much higher than that in the bulk of the gel.5 Notably, there is negligible accumulation of HA within the bulk of the alginate gel under all conditions 285

dx.doi.org/10.1021/la504393r | Langmuir 2015, 31, 283−291

Langmuir

Article

Table 2. Species of HA and Their Concentrations in the Two-Phase System Aqueous Solution/Hydrogel in the Presence of Cd(II)a species of HA in the CdHA water/ gel system

aqueous medium, bulk

freely dispersed HA dispersed HA complexed with Cd, CdHA gel-immobilized HA, HAgb gel-immobilized HA, complexed with Cd, CdHAgb

* cHA,aq cCdHA,aq *

gel surface film

gel bulk phase

governing physicochemical features

csHA,g = ΠD,HAcHA,aq * csCdHA,g = ΠD,CdHAcCdHA,aq *

cHA,g * = ΠD,HAcHA,aq * cCdHA,g * = ΠD,CdHAcCdHA,aq *

Donnan Donnan + complexation

csHA,gb = fsHA ̅ ΠD,HAc*HA,aq csCdHA,gb = fsCdHA ̅ ΠD,CdHAcCdHA,aq *

c*HA,gb = f *HAΠD,HAc*HA,aq cCdHA,gb * = f CdHA * ΠD,CdHAcCdHA,aq *

Donnan + binding by gel backbone Donnan + complexation + binding by gel backbone

a

It is assumed that the metal to ligand ratio is sufficiently small so that complexation by Cd2+ does not affect the partitioning behavior of HA. For * > 1; alginate gel, ΠD,HA < 1 and f HA * ≪ 1. fsi̅ denotes the average gel accumulation factor for species i both gels, fSHA ̅ ≫ 1; PAAm gel, ΠD,HA = 1 and f HA in the surface film.

considered herein. On the other hand, the bulk of the PAAm gel generally equilibrates to a HA concentration significantly greater than that in the aqueous medium. Here, we consider the influence of ionic strength and the presence of metal ions on the HA accumulation process. Additionally, in the case of alginate gels that contain protonatable functional groups, the influence of pH is also considered. The equilibrium concentration profiles reflect the eventual result of the combined partition/interaction features of the various particulate HA species and those of the gel as a function of the physicochemical conditions in the two phases. Table 2 summarizes the factors that govern the equilibrium distribution of HA species within the aqueous/hydrogel systems, i.e., (i) the Donnan partitioning coefficient, ΠD; (ii) the gel accumulation factor f i, defined as (ci,gb + ci,g)/ci,g, in the pertaining gel phase, which is related to the thermodynamic association constant for binding of i by the gel; and (iii) the extent to which HA is complexed with metal ions. As noted above, these parameters are expected to be very different for the two types of gel: in the bulk of the gel, PAAm is characterized by ΠD,HA = 1 and f *HA > 1, whereas for alginate, ΠD,HA < 1 and f *HA ≪ 1. Effect of Ionic Strength on Water/Gel Partitioning of HA. The influence of ionic strength on the equilibrium partitioning of humic acids between aqueous medium and the two types of gel phases is presented in Figure 2. Under the given experimental conditions, PAAm is practically uncharged, whereas alginate carries a negative structural charge density of −5.6 mol m−3.40 In the absence of Cd(II), the extent to which HA accumulates in the surface film of the alginate gel was similar for ionic strengths of 1 and 10 mol m−3 (Figure 2a). At sufficiently low ionic strength, the hydrophobic regions of the HA may be more free to interact with the gel backbone because formation of micelle-like structures is hindered by intramolecular electrostatic repulsion between the negative charges on the HA.41 However, in the case of the negatively charged alginate gel, this aspect is apparently not sufficient to overcome the repulsion between the negative charges on the HA and the gel. Furthermore, the increased extent of charge screening at an electrolyte concentration of 10 mol m−3 is not effective in reducing electrostatic repulsion between HA particles and the gel backbone. Still, there is substantial HA accumulation into the surface film, suggesting that in this particular zone hydrophobic interactions between core elements of HA entities and apolar elements of the gel backbone prevail41 over electrostatic effects. In all cases studied herein, the charge brought to the gel phase by the accumulated HA particles is small relative to the negative structural charge density of the alginate backbone. For example, at the highest accumulated HA concentration of ca. 0.3 kg m−3 at the gel/water interphase

Figure 2. Impact of ionic strength and presence of cadmium ions on HA concentration profiles in the (a) alginate and (b) PAAm gel phase upon exposure to an aqueous HA dispersion with concentration, c*HA,aq, * = 0 or 10−2 mol m−3. cHA,limit denotes the of 0.04 kg m−3, pH 6, cCd(II) highest concentration of HA that can be measured by CLSM. The gel/ water interface is at distance = 0.

(Figure 2a), the HA contributes ca. −0.7 mol m−3 of charge density, which shifts the Donnan potential at an ionic strength of 1 mol m−3 from −44 to −47 mV and correspondingly increases the gel/water partitioning factor for a divalent cation from 33 to 42. In contrast, ionic strength had a significant impact on the extent to which HA partitions into PAAm gel: the amount of HA accumulated decreases with increasing ionic strength (Figure 2b). Specifically, the average concentration of HA 286

dx.doi.org/10.1021/la504393r | Langmuir 2015, 31, 283−291

Langmuir

Article

over the whole gel volume, cH̅ A,g, decreases from 0.2 kg m−3 at 1 mol m−3 ionic strength to 0.1 kg m−3 at 10 mol m−3 and further to 0.05 kg m−3 at 100 mol m−3. The ionic strength dependence of the HA accumulation in PAAm was observed previously41 and rationalized in terms of a substantial hydrophobic affinity of the HA for the gel backbone. That is, at sufficiently low ionic strength, intramolecular electrostatic repulsion between the negative charges on HA inhibits aggregation of its hydrophobic regions, which are then freer to interact with the polymer backbone of the gel. Since the PAAm gel backbone is practically uncharged, there is negligible electrostatic repulsion between HA particles and the gel. In this context, it is relevant to note that the effective charge density within the PAAm gel is modified by the accumulated HA. The highest concentration of HA measured at the gel/water interphase at an ionic strength of 1 mol m−3, ca. 0.7 kg m−3 (Figure 2b), corresponds to ca. −2 mol m−3 of charge density, which is significant relative to the −1.2 mol m−3 of PAAm itself.13 Thus, in the presence of such an amount of HA, the Donnan potential at an ionic strength of 1 mol m−3 is shifted from −14 to −30 mV, and the gel/water partitioning factor for a divalent cation is correspondingly increased 4-fold, from 3 to 12. Effect of Cd(II) on Water/Gel Partitioning of HA. The effect of Cd on the partitioning of HA into PAAm and alginate gels is also shown in Figure 2. As detailed above, the size distribution of the HA particles and their fluorescence intensity are not significantly altered in the presence of Cd(II) at a low metal-toligand ratio (see Experimental Section for justification of the solution composition). In the case of PAAm, the concentration of Cd2+ aq in the gel phase is not significantly enhanced over that in the aqueous medium.42 Figure 2b shows that in the presence of Cd(II) the accumulation of HA generally decreases somewhat. Although there are some discrepancies at a few spatial locations, cH̅ A,g was generally lower in the presence of Cd(II) at all ionic strengths considered. As proposed previously,40 this feature may be ascribed to a change in HA conformation upon complexation of Cd(II) so that, generally, the hydrophobic regions are less free to interact with the gel backbone. In the case of alginate gels, uncomplexed Cd2+ aq is quite strongly bound by the gel backbone, with concentrations in the gel phase being greater than those in the aqueous medium by factors between 10 and 100.43 Such bound Cd(II), however, does not significantly affect the Donnan potential of the gel.9 The significant structural charge density carried by the alginate gel makes the effect of Cd(II) on the accumulation of HA more differentiated compared to that for the PAAm case. At an ionic strength of 10 mol m−3, the alginate gel carries a modest charge (Donnan potential −7 mV), and the extent to which HA accumulates in the surface film is increased in the presence of Cd(II) (Figure 2a). This behavior is ascribed to the reduction in charge of the HA particles upon complexation with Cd(II) and the ensuing reduction of the electrostatic repulsion between HA and the gel backbone. In contrast, at the lower ionic strength of 1 mol m−3, the Donnan potential is much larger (−44 mV), which changes the balance between hydrophobic attraction and electrostatic repulsion and reduces the accumulation of HA in the presence of Cd(II). Apparently, at this low ionic strength, the reduction in charge of the HA particles upon complexation is not sufficient to moderate the electrostatic repulsion between the HA carboxylate groups and the gel backbone.

Effect of pH on Water/Alginate Gel Partitioning of HA. Since the alginate gel backbone contains protonatable functional groups and thus exhibits pH-dependent charge density,10 it is of interest to explore the effect of pH on the partitioning behavior of HA in this system. Figure 3 shows the effect of pH

Figure 3. Impact of pH on (a) equilibrium HA concentration profiles in the alginate gel phase and (b) the amount of HA sorbed per unit surface area at equilibrium (Γeq, kg m−2) in the first 15 μm of the gel phase upon exposure to an aqueous HA dispersion at a concentration, c*HA,aq, of 0.04 kg m−3, 10 mol m−3 NaNO3, c*Cd(II) = 10−2 mol m−3. The points are experimental values. In panel b, the points correspond to Γeq values (left-hand y axis), and the dashed curve is the percent of protonated HA (right-hand y axis).16 cHA,limit denotes the highest concentration of HA that can be measured by CLSM.

on the accumulation of HA in alginate gels from 10 mol m−3 ionic strength solutions in the presence of Cd(II). The absolute amount of accumulated HA increases as the pH decreases, which is ascribed to reduced electrostatic repulsion between the HA particle and the gel backbone, with both of them losing negative charge.10 Furthermore, as the net charge on the HA particle decreases with decreasing pH, its hydrophilicity is reduced, and a greater affinity for the gel backbone results. In this way, the concomitant pH dependence of the net charge on the gel and on the HA entities indirectly govern the effectiveness of the hydrophobic accumulation process. There is no significant change in the overall size distribution of the HA particles over the range of conditions shown in Figure 3 (see Table 1). The importance of the extent of HA protonation on its partitioning between water and the gel phase is highlighted by comparing the percent of protonated HA with the amount of HA sorbed within the surface film per unit surface area Γeq (kg m−2) as a function of pH (Figure 3b). The 287

dx.doi.org/10.1021/la504393r | Langmuir 2015, 31, 283−291

Langmuir

Article

water/gel interface will be on the order of 50 μm.44 Thus, the time constant for setting a steady-state diffusion layer in the aqueous solution, δ2HA/D̅ HA, is an order of magnitude shorter than that for accumulation in the surface film. In addition, the equilibrium ratio between the average cHA in the surface film and c*HA,aq in the bulk aqueous solution is on the order of 10, whereas the thickness of the surface film is a factor of ca. 3 thinner than δHA. Accordingly, the temporal dependence of the sorbed HA per unit area, ΓHA (kg m−2), approximates that of a surface accumulation process governed by the rate of steadystate diffusion in the aqueous solution.5 Assuming linear sorption (Henry isotherm), with the concentration of HA at the aqueous side of the solution/gel interface, c0HA,aq gradually increasing with time in linear proportion to ΓHA through the sorption coefficient, KHA, we obtain the relationship45

pH profile of the sorption by the gel backbone follows that of the protonation of HA in the medium. Notably, even at pH 3 (Figure 3a), where the concentration of HA in the alginate surface film approaches levels observed for PAAm (Figure 2b), no HA was detected in the bulk of the gel. At lower ionic strength (1 mol m−3), a similar pH dependence of the HA accumulation is observed (Figure S3), although the effect is moderated by the much larger Donnan potentials in the gel phase at lower ionic strength (see above), which leads to the pH in the gel phase being somewhat lower than that in the aqueous medium. Temporal Evolution of HA Accumulation. Figure 4 shows the temporal evolution of the HA accumulation profile

ΓHA(t )/Γ eq HA = 1 − exp( − t / τacc)

(1)

where Γeq HA is the eventual equilibrium ΓHA and τacc is the characteristic time constant of the accumulation process defined by τacc = δ HAKHA /D̅ HA

(2)

Figure 4b indicates that there is a tendency for the surface film to be filled more rapidly as the ionic strength increases. This trend can be rationalized by considering the relative time scales for accumulation in the surface as compared to the bulk of the gel phase. Figure 2b shows that the amount of HA accumulated in the bulk of the PAAm gel, and thus KHA, increases with decreasing ionic strength. A greater KHA means a greater τacc (eq 2) and thus a longer time scale for building-up of the surface film. Accordingly, as the ionic strength decreases, the time scale of development of the surface film more strongly overlaps with that for diffusion into the bulk gel phase. In other words, the flux of HA passing through the surface film into the bulk zone of the gel phase counts more heavily. At the highest ionic strength of 100 mol m−3, eq 1 provides a reasonable fit to the data for a τacc of 3 × 103 s and thus a KHA of 6 × 10−4 m (dashed curve in Figure 4b). The initial data lie above the computed steady-state curve due to the transient nature of the diffusive HA flux in the initial short time domain, as detailed previously.5 When the surface layer accumulation approaches equilibrium, the slower process of penetration into the core of the bulk PAAm gel phase gradually starts to control the timing of the HA partitioning process. The extent to which the time scales for these processes overlap depends on the ionic strength and magnitude of KHA (see above). The HA accumulated in the surface film will effectively maintain a finite concentration gradient over the time scale of the bulk equilibration. For the gel layers used herein, with thicknesses on the millimeter scale and both sides exposed to the medium, the applicable time scale is that for nonconvective diffusion over distances of half the thickness of the gel layer, dg, i.e., τacc = (dg/2)2/D̅ HA. The characteristic accumulation time constant is further enlarged by the bulk gel accumulation factor. An average diffusion coefficient D̅ HA for HA dispersions at different electrolyte concentrations can be calculated from the Stokes−Einstein equation together with the size data given in Table 1, to yield D̅ HA equal to 7 × 10−12, 6 × 10−12, and 2 × 10−12 m2 s−1 at ionic strengths of 1, 10, and 100 mol m−3, respectively. Using these values yields a τacc that increases from 2 × 104 s at an ionic strength of 1 mol m−3 to 9 × 104 s at 100 mol m−3. The experimental data are in order-of-magnitude agreement with

Figure 4. Time evolution of accumulation of HA in PAAm gel upon exposure to an aqueous HA dispersion of 0.04 kg m−3, pH 6, in the presence of 10−2 mol m−3 Cd(II). (a) Concentration profiles at 1 mol m−3 ionic strength as a function of distance from the gel/water interface for different accumulation times; (b) temporal filling up of the surface film for various ionic strengths (points), together with the computed steady-state curve, eq 1, with τacc = 3 × 103 s (black dashed curve).

in PAAm gel for a bulk medium HA concentration of 0.04 kg m−3 (pH 6, I = 1 mol m−3, 10−2 mol m−3 Cd(II)). The time constant of accumulation in the surface film (Figure 4b) is on the order of thousands of seconds. For the conditions herein of mild solution convection with HA particles having an aqueous diffusion coefficient of 1 × 10‑11 m2 s−1, the thickness of the steady-state planar diffusion layer, δHA, at the macroscopic 288

dx.doi.org/10.1021/la504393r | Langmuir 2015, 31, 283−291

Langmuir

Article

function relationships in biogels. That is, characterization of the partitioning behavior of different types of organic matter (including humic and fulvic substances of soil and aquatic origin and size fractions thereof) into a series of gels with controlled hydrophobicity and porosity will identify the physicochemical properties of the gel phase and of the sorbing species that govern the accumulation and stratification processes. Such information would allow for rigorous understanding of the relationship between physicochemical speciation of compounds in the exposure medium and their bioavailability, which inherently involves partitioning and diffusion of bioactive species into gel layers.

this prediction (data not shown), confirming that the effective diffusion coefficient of HA in the gel phase is comparable to that in aqueous solution. Overall, diffusion is seen to be the rate-limiting step over the entire process of HA partitioning between aqueous solution and gel phases. This confirms that the rate of association of HA with the gel backbone is fast compared to the rate of diffusive supply and that local sorption equilibrium is maintained on the time scale of the overall accumulation process.

■

CONCLUSIONS HA particles are found to accumulate in PAAm and alginate gels to an extent that depends on the pH and ionic strength of the medium. Under all conditions, significantly more HA is accumulated in the PAAm gel, which is ascribed to the negligible electrostatic repulsion between HA and the practically uncharged polymer backbone. For both gels, under all conditions considered, the accumulation is spatially heterogeneous, with formation of a distinctive surface film with strong accumulation of HA. Despite the changes in hydrophilicity of HA and the gels, especially alginate, as a function of pH and ionic strength, the thickness of the surface accumulation film remains remarkably constant at ca. 15 μm. The constancy of the surface film thickness suggests that its dimensions are determined predominantly by the properties of the HA rather than those of the gels. It is hypothesized that upon entering the gel phase the local HA concentration increases due to sorption on the gel backbone, leading to growth of aggregates, which eventually reach a critical size within the constraints of the gel matrix. Nevertheless, although the concentrations of HA accumulated in the gel surface film are rather high, they appear to be well below saturation of the gel. A concentration of 0.6 kg m−3 attained in the surface of the PAAm gel corresponds to ca. 0.04% of the local volume being occupied by HA (for an HA density of ca. 1.5).46 Although less accumulation was generally observed in case of alginate, the surface accumulation seems to block further penetration of HA into the bulk gel matrix, perhaps caused by cogelation between HA and the flexible alginate cross-links.47 Further studies in the higher HA concentration range may prove to be useful and to possibly identify maximum concentrations of sorbed HA. The present results are relevant for understanding the partitioning of HA into soft interphases, the mechanisms of which underpin issues such as sink/source functioning of biogels in environmental systems, bioavailability of chemical species, and magnitudes of fluxes at sensors that employ gel layers, e.g., DGT, 12,48 and gel-coated microelectrodes (GIME).49 Furthermore, the behavior of trace compounds that associate with HA will be concomitantly affected by the accumulation processes discussed herein. As a consequence of the modified Donnan potential, the gel/water partitioning behavior of ionic compounds that do not associate with HA may also be influenced by the presence of sorbed HA. Accordingly, the influence of the type and concentration of HA in the exposure medium on the spatial distribution of, e.g., metal ions and trace organics within gel phases merits attention. In this respect, it would be useful to further detail the work and to, e.g., include a wide range of metal to HA complexing ligand concentration ratios at a range of ionic strengths to systematically change the charge density of the HA species and of the gel phase at constant pH. The results presented herein provide a basis for exploiting hydrogels as speciation controllers and elucidating structure/

■

ASSOCIATED CONTENT

* Supporting Information S

Figure S1: Calibration curve for determining the concentration of HA dispersions by CLSM. Figure S2: Impact of ionic strength, pH, and presence of cadmium ions on the fluorescence intensity of an aqueous HA dispersion. Figure S3: Impact of pH on HA concentration profiles in the alginate gel phase at an ionic strength of 1 mol m−3. This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work was performed within the framework of the BIOMONAR project funded by the European Commission’s seventh framework program (Theme 2: Food, Agriculture and Biotechnology), under grant agreement 244405.

■

289

ABBREVIATIONS CLSM, confocal laser scanning microscopy DLS, dynamic light scattering HA, humic acid PAAm, modified polyacrylamide cCd(II) * , total concentration of Cd(II) in the bulk aqueous solution (mol m−3) c*HA,aq, concentration of HA in the bulk aqueous solution (kg m−3) c0HA,aq, concentration of HA at the aqueous side of the solution/gel interface (kg m−3) csHA,g, concentration of HA in the gel liquid of the surface film (kg m−3) csHA,gb, concentration of gel-bound HA in the surface film (kg m−3) c*HA,g, concentration of HA in the bulk of the gel phase (kg m−3) cHA,gb * , concentration of gel-bound HA in the bulk of the gel phase (kg m−3) cH̅ A,g, average eventual equilibrium concentration of HA in the gel layer (kg m−3) D̅ HA, average diffusion coefficient of HA (m2 s−1) dg, thickness of gel layer (m) δHA, diffusion layer thickness for HA in aqueous solution (m) f i, gel accumulation factor of species i ΓHA, sorbed amount of HA per unit surface area (kg m−2) dx.doi.org/10.1021/la504393r | Langmuir 2015, 31, 283−291

Langmuir

Article

−2 Γeq HA, ΓHA at equilibrium (kg m ) −3 I, ionic strength (mol m ) KHA, sorption coefficient for HA (m) ΠD,i, Donnan partitioning coefficient of species i rp, particle radius (nm) rpeak, radius at the peak of the particle size distribution (nm) t, time (s) τacc, time constant for accumulation (s)

■

(19) Pinheiro, J. P.; Mota, A. M.; d’Oliveira, J. M. R.; Martinho, J. M. G. Dynamic properties of humic matter by dynamic light scattering and voltammetry. Anal. Chim. Acta 1996, 329, 15−24. (20) Kučerík, J.; Šmejkalová, D.; Č echlovská, H.; Pekař, M. New insights into aggregation and conformational behavior of humic substances: application of high resolution ultrasonic spectroscopy. Org. Geochem. 2007, 38, 2098−2110. (21) Guetzloff, T. F.; Rice, J. A. Does humic acid form a micelle? Sci. Total Environ. 1994, 152, 31−35. (22) von Wandruszka, R. The micellar model of humic acid: evidence from pyrene fluorescence measurements. Soil Sci. 1998, 163, 921−930. (23) Kerner, M.; Hohenberg, H.; Ertl, S.; Reckermann, M.; Spitzy, A. Self-organization of dissolved organic matter to micelle-like microparticles in river water. Nature 2003, 422, 150−154. (24) Young, C.; von Wandruszka, R. A comparison of aggregation behavior in aqueous humic acids. Geochem. Trans. 2001, 2, 16−21. (25) Bryan, N. D.; Jones, M. N.; Birkett, J.; Livens, F. R. Application of a new method of analysis of ultracentrifugation data to the aggregation of a humic acid by copper(II) ions. Anal. Chim. Acta 2001, 437, 281−289. (26) Palmer, N. E.; von Wandruska, R. The influence of aggregation on the redox chemistry of humic substances. Environ. Chem. 2009, 6, 178−184. (27) Balnois, E.; Wilkinson, K. J.; Lead, J. R.; Buffle, J. Atomic force microscopy of humic substances: effects of pH and ionic strength. Environ. Sci. Technol. 1999, 33, 3911−3917. (28) Trubetskaya, O.; Trubetskoj, O.; Guyot, G.; Andreux, F.; Richard, C. Fluorescence of soil humic acids and their fractions obtained by tandem size exclusion chromatography−polyacrylamide gel electrophoresis. Org. Geochem. 2002, 33, 213−220. (29) Mobed, J. J.; Hemmingsen, S. L.; Autry, J. L.; McGown, L. B. Fluorescence characterization of IHSS humic substances: total luminescence spectra with absorbance correction. Environ. Sci. Technol. 1996, 30, 3061−3065. (30) Halim, M.; Spaccini, R.; Parlanti, E.; Amezghai, A.; Piccolo, A. Differences in fluorescence properties between humic acid and its size fractions separated by preparative HPSEC. J. Geochem. Explor. 2013, 129, 23−27. (31) Richard, C.; Trubetkaya, O.; Trubetkoj, O.; Reznikova, O.; Afanaseva, G.; Aguer, J.- P.; Guyot, G. Key role of the low molecularsize fraction of soil humic acids for fluorescence and photoinductive activity. Environ. Sci. Technol. 2005, 38, 2052−2057. (32) Boyle, E. S.; Guerriero, N.; Thiallet, A.; del Vecchio, R.; Blough, N. V. Optical properties of humic substances and CDOM: relation to structure. Environ. Sci. Technol. 2009, 43, 2262−2268. (33) Conte, P.; Spaccini, R.; Šmejkalová, D.; Nebbioso, A.; Piccolo, A. Spectroscopic and conformational properties of size-fractions separated from a lignite humic acid. Chemosphere 2007, 69, 1032− 1039. (34) Plaza, C.; Brunetti, G.; Senesi, N.; Polo, A. Molecular and quantitative analysis of metal ion binding to humic acids from sewage sludge and sludge-amended soils by fluorescence spectroscopy. Environ. Sci. Technol. 2006, 40, 917−923. (35) Provenzano, M. R.; D’Orazio, V.; Jerzykiewicz, M.; Senesi, N. Fluorescence behaviour of Zn and Ni complexes of humic acids from different sources. Chemosphere 2004, 55, 885−892. (36) Plaza, C.; Brunetti, G.; Senesi, N.; Polo, A. Molecular and quantitative analysis of metal ion binding to humic acids from sewage sludge and sludge-amended soils by fluorescence spectroscopy. Environ. Sci. Technol. 2006, 40, 917−923. (37) Terbouche, A.; Djebbar, S.; Benali-Baitich, O.; Bouet, G. Characterization and complexing capacity of humic acid extracted from Yakouren soil with heavy metals by conductimetry and quenching of fluorescence. Soil Sediment Contamin. 2010, 19, 21−41. (38) Riggle, J.; von Wandruszka, R. Dynamic conductivity measurements in humic and fulvic acid solutions. Talanta 2004, 62, 103−108. (39) Pullin, M. J.; Cabaniss, S. E. Rank analysis of the pH-dependent synchronous fluorescence spectra of six standard humic substances. Environ. Sci. Technol. 1995, 29, 1460−1467.

REFERENCES

(1) Liu, A. G.; Wu, R. C.; Eschenazi, E.; Papadopoulos, K. AFM on humic acid adsorption on mica. Colloids Surf., A 2000, 174, 245−252. (2) Plaschke, M.; Romer, J.; Klenze, R.; Kim, J. I. In situ AFM study of sorbed humic acid colloids at different pH. Colloids Surf., A 1999, 160, 269−279. (3) Bayen, S.; ter Laak, T. L.; Buffle, J.; Hermens, J. L. M. Dynamic exposure of organisms and passive samplers to hydrophobic chemicals. Environ. Sci. Technol. 2009, 43, 2206−2215. (4) Wilkinson, K. J.; Buffle, J. Critical evaluation of the physicochemical parameters and processes for modeling the biological uptake of trace metals in environmental (aquatic) systems. In Physicochemical Kinetics and Transport at Biointerfaces; van Leeuwen, H. P., Köster, W., Eds.; Wiley: Hoboken, NJ, 2004; Vol. 9, pp 445− 533. (5) Zielińska, K.; Town, R. M.; Yasadi, K.; van Leeuwen, H. P. Partitioning of humic acids between aqueous solution and hydrogel. Concentration profiling of humic acids in hydrogel phases. Langmuir 2014, 30, 2084−2092. (6) Cogan, N. G.; Keener, J. P. The role of the biofilm matrix in structural development. Math. Med. Biol. 2004, 21, 147−166. (7) Haug, A.; Smidsrød, O. The effect of divalent metals on the properties of alginate solutions. II. Comparison of different metal ions. Acta Chem. Scand. 1965, 19, 341−351. (8) Davis, T. A.; Llanes, F.; Volesky, B.; Mucci, A. Metal selectivity of Sargassum spp. and their alginates in relation to their a-L-guluronic acid content and conformation. Environ. Sci. Technol. 2003, 37, 261−267. (9) Kalis, E. J. J.; Davis, T. A.; Town, R. M.; van Leeuwen, H. P. Impact of ionic strength on Cd(II) partitioning between alginate gel and aqueous media. Environ. Sci. Technol. 2009, 43, 1091−1096. (10) Kalis, E. J. J.; Davis, T. A.; Town, R. M.; van Leeuwen, H. P. Impact of pH on CdII partitioning between alginate gel and aqueous media. Environ. Chem. 2009, 6, 305−310. (11) Secchi, E.; Roversi, T.; Buzzaccaro, S.; Piazza, L.; Piazza, R. Biopolymer gels with “physical” cross-links: gelation kinetics, aging, heterogeneous dynamics, and macroscopic mechanical properties. Soft Matter 2013, 9, 3931−3944. (12) Davison, W.; Zhang, H. Progress in understanding the use of diffusive gradients in thin films (DGT)back to basics. Environ. Chem. 2012, 9, 1−13. (13) Yezek, L.P.; van Leeuwen, H.P. An electrokinetic characterization of low charge density cross-linked polyacrylamide gels. J. Colloid Interface Sci. 2004, 278, 243−250. (14) van der Veeken, P. L. R.; Pinheiro, J. P.; van Leeuwen, H. P. Metal speciation by DGT/DET in colloidal complex systems. Environ. Sci. Technol. 2008, 42, 8835−8840. (15) Laca, A.; Garcia, L. A.; Argüeso, F.; Diaz, M. Protein diffusion in alginate beads monitored by confocal microscopy. The application of wavelets for data reconstruction and analysis. J. Ind. Microbiol. Biotechnol. 1999, 23, 155−165. (16) Avena, M. J.; Koopal, L. K.; van Riemsdijk, W. H. Proton binding to humic acids: electrostatic and intrinsic interactions. J. Colloid Interface Sci. 1999, 217, 37−48. (17) Davison, W.; Zhang, H.; Grime, G. W. Performance characteristics of gel probes used for measuring the chemistry of pore waters. Environ. Sci. Technol. 1994, 28, 1623−1632. (18) Buffle, J. Complexation Reactions in Aquatic Systems: An Analytical Approach; Ellis Horwood: Chichester, 1988. 290

dx.doi.org/10.1021/la504393r | Langmuir 2015, 31, 283−291

Langmuir

Article

(40) van der Veeken, P. L. R.; van Leeuwen, H. P. DGT/DET gel partition features of humic acid/metal species. Environ. Sci. Technol. 2010, 44, 5523−5527. (41) van der Veeken, P. L. R.; Chakraborty, P.; van Leeuwen, H. P. Accumulation of humic acid in DET/DGT gels. Environ. Sci. Technol. 2010, 44, 4253−4257. (42) Yezek, L. P.; van Leeuwen, H. P. Donnan effects in the steadystate diffusion of metal ions through charged thin films. Langmuir 2005, 21, 10342−10347. (43) Davis, T. A.; Kalis, E. J. J.; Pinheiro, J. P.; Town, R. M.; van Leeuwen, H. P. Cd(II) speciation in alginate gels. Environ. Sci. Technol. 2008, 42, 7242−7247. (44) Crank, J. The Mathematics of Diffusion, 2nd ed.; Oxford University Press: Oxford, 1973. (45) Bard, A. J.; Faulkner, L. R. Electrochemical Methods: Fundamentals and Applications; 2nd ed.; Wiley: New York, 2001. (46) Birkett, J. W.; Jones, M. N.; Bryan, N. D.; Livens, F. R. Computer modeling of partial specific volumes of humic substances. Eur. J. Soil Sci. 1997, 48, 131−137. (47) Morris, E. R. Mixed polymer gels. In Food Gels; Harris, P., Ed.; Elsevier: London, 1990; pp 291−359. (48) van Leeuwen, H. P. Steady-state DGT fluxes of nanoparticulate metal complexes. Environ. Chem. 2011, 8, 525−528. (49) Belmont-Hebert, C.; Tercier, M. L.; Buffle, J.; Fiaccabrino, G. C.; de Rooij, N. F.; Koudelka-Hep, M. Gel-integrated microelectrode arrays for direct voltammetric measurements of heavy metals in natural waters and other complex media. Anal. Chem. 1998, 70, 2949−2956.

291

dx.doi.org/10.1021/la504393r | Langmuir 2015, 31, 283−291