Speciation Analysis of Aqueous Nanoparticulate Diclofenac

Sep 18, 2012 - The surface concentration of bound X, ΓX–S, and the total concentration of the binding sites, ΓS,max, can be translated to the corr...
0 downloads 0 Views 883KB Size
Article pubs.acs.org/Langmuir

Speciation Analysis of Aqueous Nanoparticulate Diclofenac Complexes by Solid-Phase Microextraction Katarzyna Zielińska,† Herman P. van Leeuwen,† Sylvain Thibault,‡ and Raewyn M. Town*,‡ †

Laboratory of Physical Chemistry and Colloid Science, Wageningen University, Dreijenplein 6, 6703 HB Wageningen, The Netherlands ‡ Institute for Physics, Chemistry and Pharmacy, University of Southern Denmark, Campusvej 55, DK-5230 Odense, Denmark S Supporting Information *

ABSTRACT: The dynamic sorption of an organic compound by nanoparticles (NPs) is analyzed by solid-phase microextraction (SPME) for the example case of the pharmaceutical diclofenac in dispersions of impermeable (silica, SiO2) and permeable (bovine serum albumin, BSA) NPs. It is shown that only the protonated neutral form of diclofenac is accumulated in the solid phase, and hence this species governs the eventual partition equilibrium. On the other hand, the rate of the solid/water partition equilibration is enhanced in the presence of the sorbing nanoparticles of SiO2 and BSA. This feature demonstrates that the NPs themselves do not enter the solid phase to any appreciable extent. The enhanced rate of attainment of equilibrium is due to a shuttletype of contribution from the NP-species to the diffusive supply of diclofenac to the water/solid interface. For both types of nanoparticulate complexes, the rate constant for desorption (kdes) of bound diclofenac was derived from the measured thermodynamic affinity constant and a diffusion-limited rate of adsorption. The computed kdes values were found to be sufficiently high to render the NP-bound species labile on the effective time scale of SPME. In agreement with theoretical prediction, the experimental results are quantitatively described by fully labile behavior of the diclofenac/ nanoparticle system and an ensuing accumulation rate controlled by the coupled diffusion of neutral, deprotonated, and NPbound diclofenac species.



INTRODUCTION Organic micropollutants, such as pharmaceuticals, have attracted considerable attention in recent decades.1−4 The biological and environmental fate and reactivity of organic pollutants are highly dependent on their chemical forms, yet their possible interactions with other components of aquatic ecosystems, for example, humic substances, proteins, or engineered nanoparticles, remain largely unknown. There is thus an urgent need to develop experimental tools and conceptual frameworks to quantitatively describe, and predict, the speciation of organic micropollutants in a wide range of environmental and biological samples. Solid-phase microextraction (SPME) is a useful tool for speciation analysis of organic micropollutants. SPME was developed in the late 1980s as a solvent-free and relatively fast technique5 and has subsequently gained widespread acceptance for selective extraction of target compounds prior to their separation and determination, often using some form of chromatography.6−8 SPME is based on the partitioning of the target compounds between the sample and a solid polymeric phase, which can take place in the sample headspace or directly in the sample. Particles in the nanosized range (nanoparticles, NPs) are ubiquitous in the environment and encompass both permeable and impermeable entities, for example, enzymes, protein molecules, clays, and soot. In addition to natural NPs, a number of engineered NPs have recently been developed and © 2012 American Chemical Society

are attracting a lot of attention due to their unique properties and widespread application. However, their reactivities and fluxes in environmental compartments and consequent risk posed to biological systems are still largely unknown.9 In the aquatic environment, natural and engineered nanoparticles are known to impact on the binding, transport, and fate of vital and toxic compounds such as organics and heavy metals.10,11 Depending on the system, both enhancement and reduction in the mobility, bioavailability, and degradation kinetics of organics in the presence of NPs have been reported.12−15 To date, the speciation of metals is relatively well investigated;16,17 however, there is a paucity of studies that address the speciation of organic compounds in aquatic systems,18 in particular in the presence of dispersions of potentially sorbing colloidal particles.8,19 The focus of the present study is to explore the dynamic features of an organic target compound in the presence of sorbing NPs. Specifically, we use SPME to characterize the chemodynamics of diclofenac with two model NPs, silica (impermeable) and bovine serum albumin, BSA (permeable). Diclofenac is a nonsteroidal antiinflammatory drug that is widely prescribed in human and veterinary medicine and is recalcitrant to standard wastewater treatment.20,21 Furthermore, in common with many emerging Received: August 3, 2012 Revised: September 16, 2012 Published: September 18, 2012 14672

dx.doi.org/10.1021/la303143w | Langmuir 2012, 28, 14672−14680

Langmuir

Article

aqueous solution are very fast;22,23 thus the various protonated forms of X will be in equilibrium, and their rates of interconversion will not affect overall sorption/desorption rates. In the following, unless stated otherwise, we use the term “free” to denote the analyte fraction that is unbound to NPs and collectively represents its various protonated forms. Similarly X−S collectively denotes the bound analyte irrespective of its protonation state or that of the sorbent. In general, we assume that the final attachment step in the adsorption of HxX at/within the NP is fast (step 2 in eq 2). Accordingly, the overall rate of adsorption, Rad, can be estimated from the rate of diffusive supply of the organic molecule to the spherical NP surface, via an approach analogous to that developed for metal binding by nanoparticulate complexants.24 For the case of a neutral organic compound and a charged NP, the electric field has no impact on the kinetics,25 and thus the mere diffusive flux expression yields the reaction rate for adsorption from bulk solution (Rad, expressed as −dcw,X * /dt):

organic pollutants, it exists in both protonated (neutral) and deprotonated forms in the ambient pH range. Generally the neutral form is the relevant species for the partitioning of diclofenac between an aqueous phase and a more hydrophobic phase. The impact of unprotonated species on the dynamics of SPME extraction has not been previously addressed. Here, we shall analyze the dynamic features of the diclofenac/NP system under the conditions of an SPME experiment. This will include aspects such as the impact of charged (unprotonated) diclofenac species and the lability of the nanoparticulate diclofenac complex.



THEORY Equilibrium Properties of the Nanoparticulate Complex. If all binding sites of a nanoparticulate sorbent have the same affinity for a sorbate molecule X, then at low degrees of surface coverage we can assume linear sorption behavior (Henry regime), that is: *X ΓX − S = KHc w,

[mol m−2]

(1)

*X R ad = 4πrpD X NAvc p*c w,

where KH is the Henry coefficient (m). The concentration of the free, unsorbed form is usually given in terms of the bulk volume concentration (c*w,X, mol m−3), and the bound form is typically expressed as a local surface concentration (ΓX−S, mol m−2) for impermeable particles, or a local volume concentration within the particle (cX−S, mol m−3) for permeable particles. The surface concentration of bound X, ΓX−S, and the total concentration of the binding sites, ΓS,max, can be translated to the corresponding smeared-out volume concentrations by multiplication by (ANP/V), where V is the volume of the dispersion and ANP (=4πr2pc*p NAv) is the total surface area of the supposedly spherical particles, where rp is the particle radius. For permeable NPs, cX−S replaces ΓX−S, and cS,max replaces ΓS,max (and ANP/V becomes VNPNNP/V, where VNP is the NP volume and NNP is the number of NPs (=c*p NAv)). The thermodynamic stability constant of the simple Langmuirian NP-X complex KX (m3 mol−1) can be derived from KH and ΓS,max (KX = KH/ ΓS,max). Dynamic Nature of the Complexation Processes in Bulk Medium. We consider samples in which the organic target molecule (X) can be freely dissolved, in various protonated forms, and can undergo a physicochemical association reaction with a site S, located either on the surface or within the body of a nanoparticle, to yield a nanoparticulate complex:

[mol m−3 s−1]

(3a)

where c*p is the concentration of particles per unit volume of sample dispersion. When expressed per site, Rad can be expressed in kad: * X = kadcS*c w, * X = kad *X ′ c w, R ad = 4πrpD X NAvc p*NSc w, [mol m−3 s−1]

(3b)

where cS* = cp*NS, and NS is the number of reactive sites per NP. Equation 3 holds for rp ≫ rX and Dp ≪ DX. The reaction rate for desorption from the spherical NPs (Rdes, expressed as dΓX−S/dt or dcX−S/dt) is: R des = kdes ΓX − S

[mol m−2 s−1]

(4)

Together with the thermodynamic stability constant, K, of the NP−X complex, the kad computed from eq 3b allows derivation of kdes, and consequent prediction of the dynamic behavior of the NP-bound organic for a given spatial scale and time scale (see section on lability). In terms of a smeared-out volume reaction, the system shown in eq 2 can be dynamically classified on the basis of the rate constants kad and kdes. The various X species in eq 2 each have characteristic lifetimes that derive from these rate constants: for X it is 1/k′ad, and for X−S it is 1/kdes. A system is denoted as dynamic if the operational time scale, t, is much larger than the lifetimes of X and X−S, that is: ′t kad

and

kdest ≫ 1

(5)

At the other extreme, a system is static (or inert) when the lifetimes of X and X−S are much greater than t: ′t kad

and

kdest ≪ 1

(6)

Equations 4 and 5 describe the ability of a system to attain bulk equilibrium within a certain time, t. Lability of Nanoparticulate Complexes (in a Macroscopic Surface Process). We consider the dynamic nature of the interfacial SPME accumulation process for the case where the nanoparticulate complexes do not enter the solid phase. During an SPME experiment involving accumulation of the target organic X from a sample containing sorbing nanoparticles, X will partition between three phases: solid phase, aqueous solution, and bound by the nanoparticles. In this

where a is 0 to n, and x is the degree of protonation of the sorbed form of X, kdif is the rate constant for diffusive supply of X from the bulk medium to the spherical NP (see below), and kad and kdes are the respective overall rate constants for adsorption and desorption of X. The ratio kad/kdes is known as the thermodynamic adsorption constant, K. In case of sufficient excess of sites over organic molecules, the product of kad and the smeared-out volume concentration of sites kadcS (=kad ′ ) is essentially constant. Protonation/deprotonation reactions in 14673

dx.doi.org/10.1021/la303143w | Langmuir 2012, 28, 14672−14680

Langmuir

Article

3 ≫ 1, the system is labile (and thus diffusion-controlled), while for 3 ≪ 1, kdes determines the extent to which X−S contributes to the accumulation flux. In the case of a desorption rate-limited supply of X, the flux can be expressed as:

situation, the rates of transport of organic species into the solid phase may differ from those for a blank solution in which only free X is present (see below). Depending on the properties (charge/polarity) of both the solid phase and the protonated forms of X, it may be that only specific types of species are accumulated, for example, the neutral form X0. In case of binding of the target analyte to NPs in the sample, the extent to which bound species, X−S, contribute to the overall flux of X, J, depends on their lability. The SPME steadystate flux concentration profiles are shown schematically in Figure 1 for different kinetic situations.8,26 The magnitude of J

Jkin =

kdes ΓX − SANPμ V

(7)

where μ represents the layer of solution adjacent to the solidphase interface, where the sorption equilibrium between free and bound organic molecules is distorted (usually denoted as the “reaction layer”).29 The magnitude of μ depends on the mobility of the free molecule (represented by the diffusion coefficient, Dw,X) and its mean free lifetime (1/k′ad):29 ′ )1/2 μ = (Dw, X /kad

(8)

In the presence of various protonated species, μ is determined by the weighted contributions to k′ad of all of these species because they maintain protolytic equilibrium with the partitioning neutral species.23 In a fully labile sorptive system, the diffusive flux is governed by the coupled diffusion of free and bound analyte:

* = Jdif

Dc̅ t,*X (9)

δ̅

where c*t,X is the total concentration of X in the sample (c*w,X + c*X−S), δ̅ is the thickness of the joint steady-state diffusion layer as derived from the mean diffusion coefficient, D̅ . The latter is defined as the weighted average of the diffusion coefficients of X and X−S:

Figure 1. Sketch of steady-state concentration profiles of the accumulating species X0 (c*w,X0) and nanoparticulate-bound X (cX̅ −S) for the inert case (− · − · −), the labile case (− − −), and the case controlled by desorption kinetics (· · ·). c0s,X is the concentration of X at the water−solid interphase, cs̅ ,X is the mean concentration of X in the solid phase, ceq s,X denotes the final equilibrium concentration of X in the bulk solid phase, ds is the thickness of the solid phase, δ is the diffusion layer thickness, and μ is the reaction layer thickness. The various protonated forms of X all follow the profile of X0 in proportion, but are not shown for clarity (see text for details).

D̅ =

* X + D X − S ΓX − SANP /V Dw, X c w, ct,*X

(10)

where DX−S is the diffusion coefficient of the adsorbed X, which, for small organic compounds, is approximately the same as that for the nanoparticle alone. Combination of eqs 7−10 yields the lability criterion, Jkin/Jdif * (=3 ) as:

depends on the mobility of all species and the kinetics of the adsorption/desorption reactions in eq 2. Two limiting situations can be identified. First, the nonlabile case arises when kdes is sufficiently small so that the contribution of X−S species to the supply of X to the solid phase is governed by desorption kinetics, rather than by diffusion. In the opposite situation, the labile case, rates of adsorption/desorption are so fast (k′adt, kdest ≫ 1) that equilibrium is maintained between X and X−S on all relevant spatial scales and time scales.27 In a labile system, the magnitude of the flux of X to the solid/water interface is determined by the coupled diffusion of X and X−S, as determined by their respective concentrations and mobilities. In this situation, the nanoparticles merely act as carriers for the transport of X toward the solid phase/medium interface. We consider the case where SPME accumulation takes place under conditions of steady-state diffusion in the medium and in the absence of bulk depletion. Accordingly, the relevant time scale for lability is the characteristic time τss of establishing steady state, that is, τss = δ2/Dw,X. For the experimental conditions herein, τss for the diffusion of diclofenac (diffusion coefficient Dw,X = 3.9 × 10−10 m2 s−1,28 diffusion layer thickness δ ≈ 50 μm) is approximately 5 s. The lability of a complex species is defined by the relative magnitudes of the kinetic flux, Jkin (i.e., the flux corresponding to the rate of release of X from the particles), and its purely * . The ratio Jkin/Jdif * is often denoted by 3 . For diffusive flux, Jdif

3=

′ )1/2 kdes ΓX − S(ANP /V )(Dw, X /kad ≫1 Dc̅ t,*X /δ ̅

(11)

where ΓX−S (ANP/V)/ct,X * may also be written as KX′ /(KX′ + 1). Extraction Kinetics. Accumulation of X from Noncomplexing Media. We consider the usual case in which the neutral form, X0, is the relevant species for equilibrium partitioning with the more hydrophobic solid phase. If the extraction process is nondepletive, that is, the bulk concentration of the neutral free form that is extracted, c*w,X0, is essentially unaffected by the steady-state accumulation process, then the average concentration of X within the SPME layer (cs̅ ,X) evolves with time (t) according to an exponential function:30,31 * X 0K sw(1 − exp−kXt ) cs,̅ X = c w,

(12)

where Ksw is the partition coefficient of the analyte between the solid phase and the aqueous solution. Many organic compounds, including diclofenac, undergo protonation/deprotonation reactions within the environmentally relevant pH range. It is the equilibrium concentration of the accumulated species that is relevant for determining Ksw, while other nonpartitioning but labile forms effectively enhance the rate of 14674

dx.doi.org/10.1021/la303143w | Langmuir 2012, 28, 14672−14680

Langmuir

Article

to 250 °C (held for 5 min). The detector was a mass spectrometer (model HP 5973 mass selective detector) under electron-impact ionization (EI, 70 eV, 200 °C) and full-scan (m/z 35−435) or selected ion-monitoring (SIM, m/z 242.2) conditions. In studies with BSA, an Agilent 19091J-413 column (HP-5, 30 m × 0.320 mm × 0.25 μm) was used in combination with a flame ionization detector (FID). The column temperature was held at 90 °C for 5 min, increased at 20 °C min−1 to 220 °C, then increased at 10 °C min−1 to 300 °C (held for 5 min). A constant flow of hydrogen (3.2 cm3 min−1) was maintained in the column. The FID makeup gas was nitrogen (45 cm3 min−1), and the detector temperature was 320 °C. SPME Measurements. Diclofenac is a fairly polar molecule with log Kow (pH 9.2−10, I = 0.1−0.15 M) in the range 0.6−0.8,34−36 and thus not too hydrophobic solid-phase polymers, PA and PEG, were selected for the analyses in the absence and presence of SiO2 and BSA NPs, respectively. The PA fibers were cut to a length of 4 cm (yielding 5.2 × 10−10 m3 of PA per sampling fiber), cleaned with GC grade methanol, and stored in ultrapure water. The PEG fibers are supplied with length 1 cm and were conditioned according to the manufacturer’s instructions. The fibers were exposed to the sample solutions under mild stirring on a rock and roller shaker (Meettech, NL) at ca. 20 °C (SiO2) or magnetic bar stirring at ca. 25 °C (BSA). Measurements were performed in duplicate. With PA fibers, the amount of diclofenac accumulated by the solid-phase film was determined by extracting the fiber with 1 cm3 of methanol (extraction recovery 100%). This extract (2 μL) was directly injected into a GC for quantification. PEG fibers were rinsed with Milli-Q water immediately after exposure, then wiped with a soft tissue and desorbed in a GC (5 min in the splitless injector port at 250 °C). Negligible depletion of the bulk solution concentrations was verified by repeating measurements for much larger sample volumes. Attempts were made to directly measure whether SiO2 sorbs onto or penetrates into the PA solid phase. However, the use of fluorescenttagged SiO2 was confounded by the fluorescence of PA itself, and attempts to measure the Si content of the solid phase were hampered by the glass core of the SPME fiber. Nevertheless, the temporal accumulation curves for DCF demonstrate unambiguously that the NP species themselves are not significantly partitioning into the solid phase: in the presence of NPs, (i) the accumulation rate of DCF is increased and (ii) the equilibrium concentration of DCF in the solid phase is decreased (see Results and Discussion).

transport of the former due to their rapid rates of interconversion in aqueous solution. In noncomplexing media, the latter species comprise the various protonated/ deprotonated charged/uncharged forms of X. In eq 12, kX is the accumulation rate constant, which for Ksw ≫ 1 and Ds,X not ≪ Dw,X is generally determined by the diffusive mass transfer in the aqueous phase:19 kX =

A sDw, X VsK swδ

(13)

where As and Vs are the surface area and volume of the solid phase, respectively. Here, we assume laminar convection for purposes of computing the diffusion layer thickness and neglect the cylindrical nature of the SPME extraction process, which has only a minor impact on the mass transfer coefficient (D/δ) of the analyte.32 Accumulation of X from Complexing Media. In the case of a system in which the analyte is present in free and NP-bound labile forms, with the nanoparticulate species X−S not partitioning between the solid and aqueous phases, the overall flux in the sample medium is governed by coupled diffusion of the two types of species, and the accumulation rate constant (as appearing in eq 12) is then given by:8 kX̅ =

A sD̅ * X /(c w, * X + c lab, * X )]δ ̅ VsK sw[c w,

(14)

where k̅X is the effective accumulation rate constant for the complex system and clab,X * is the concentration of all labile species of X in the aqueous medium. Inspection of eq 14 shows that in the presence of labile nanoparticulate X−S species, the accumulation rate constant of X will be increased relative to that for a solution containing only free X (eq 13). This is a consequence of the coupled diffusion of X and X−S in the accumulation flux (eq 9). That is, in the presence of labile, nonpartitioning species, the eventual partition equilibrium between the solid phase and the solution is attained faster than for a sample containing only free organic compound. Evidently, when X is the only species accumulated, then the accumulated amount in the solid phase is lower than that for a blank with the same total concentration of X in the absence of binding nanoparticles.





RESULTS AND DISCUSSION Affinity of Diclofenac for SiO2 and BSA NPs. The speciation data for diclofenac (DCF) in the presence of SiO2 nanoparticles are shown in Figure 2. The curve is constructed from the SPME measurements, assuming that only the free protonated DCF is accumulated in the fiber (see below). At low degrees of surface coverage, the adsorption is linear, consistent with the Henry regime. The slope of the plot in Figure 2 yields a Henry coefficient KH of 3 × 10−7 m, that is, rather weak sorption. Assuming the effective number of sorption sites per NP to be on the order of tens,37 the corresponding thermodynamic stability constant for binding of DCF to the NPs, KDCF, is estimated as 10 m3 mol−1 (KDCF = KH/ΓS,max). Serum albumin has two major types of binding sites (commonly referred to as site I and site II) that are involved in the binding of various drug molecules.38,39 A 1:1 stoichiometry of binding to BSA has been reported for a wide range of fairly small organic molecules.40 The binding of diclofenac with BSA has been measured by several workers,41−43 and a conditional log K of 1.8 (m3 mol−1) has been reported for pH 7.0 (10 mM phosphate buffer, T = 25 °C).41 This moderate strength of association between diclofenac and BSA is comparable to that reported for BSA interactions with a wide range of other small organic molecules.40 The interaction

EXPERIMENTAL SECTION

Materials. Dicolofenac (purity ≥98%), Ludox LS nanoparticles (30% (w/w) dispersion with a density of 1210 kg m−3, a mean radius of 7.5 nm, and molar mass computed as ca. 2 × 106 g mol−1), bovine serum albumin (molar mass 66 000 g mol−1, radius ca. 3.6 nm),33 methanol, nitric acid, potassium phosphate, and disodium hydrogen phosphate were obtained from Sigma-Aldrich. Sodium nitrate solution was prepared from solid NaNO3 (Merck, Suprapur). Glass fibers with a core diameter of 110 μm and a 28.5 μm polyacrylate film (PA) were obtained from Poly Micro Industries (Phoenix, AZ). Polyethylene glycol (PEG) coated fibers, with a core diameter of 130 μm and a film thickness 60 μm, were obtained from Supelco. The ultrapure deionized water (R ≥ 18 MΩ) was prepared by a Millipore water purification system, equipped with an organic-free kit (Millipore Waters, The Netherlands). Gas Chromatography. Diclofenac concentrations were determined by gas chromatography (Hewlett-Packard gas chromatograph model HP 6890). In studies with silica NPs, separations were carried out using an Agilent 190915-433 column, 30 m × 0.25 mm, 0.25 μm film thickness. Helium (99.999%) was used as a carrier gas at 1 cm3 min−1. The GC temperature program was: 2 min in 70 °C, first ramp at 10 °C min−1 to 175 °C (held for 2 min), then ramp at 10 °C min−1 14675

dx.doi.org/10.1021/la303143w | Langmuir 2012, 28, 14672−14680

Langmuir

Article

Table 1. pH Dependence of the Thermodynamic Partition Coefficient, Ksw, for Diclofenac between the Solid-Phase Polyacrylate and Aqueous Solution PA fiber pH

% of protonated diclofenac

3 4 5 6 7

95 67 17 2 0.2

a

cs̅ ,DCF/mol m−3 13.5 12.0 3.9 0.42 0.03

Ksw 5.0 6.4 8.3 7.5 6.0

× × × × ×

103 103 103 103 103

a

Calculated for log K1H = 4.31 (published value47 corrected to I = 0.001 M); total concentration of diclofenac in bulk solution = 2.8 × 10−3 mol m−3.

preferentially sorbs the protonated form of diclofenac.51 The interaction between DCF and SiO2 is proposed to involve hydrogen bonding between the surface silanol groups on SiO2 and the carboxylic acid group in DCF.51 Accordingly, the amount of NP-bound diclofenac is expected to decrease with increasing pH and decreasing ionic strength due to both the increased negative charge on SiO2 and the greater proportion of negatively charged deprotonated diclofenac. Indeed, sorption of diclofenac to soil was reported to decrease significantly from pH 4.3 to 6.8.52 The temporal profiles for the SPME accumulation of diclofenac in the absence and presence of SiO2 NPs are given in Figure 3a. The curves exhibit the expected shape of an approximately exponential accumulation function. In the absence of NPs, the fit to eq 12 yields an accumulation rate constant for diclofenac of kDCF = 2.5 × 10−5 s−1. Combining this with Ksw and the diffusion coefficient Dw,DCF yields an effective diffusion layer thickness δ of 54 μm (eq 13), which is quite reasonable in comparison with findings for atrazine with PDMS fibers.8 The temporal profiles for the SPME accumulation of diclofenac in the absence and presence of BSA are given in Figure 4. For the diclofenac only case, the fit to eq 12 yields an accumulation rate constant for diclofenac of kDCF = 1.4 × 10−5 s−1. The effective diffusion layer thickness δ is found to be 26 μm (eq 13), that is, a bit thinner than in the SiO2 experiments due to the slower stirring employed for the latter system. The eventual equilibrium concentration of diclofenac in the solid phase is given by Ksw and the concentration of the aqueous diclofenac species that accumulates in the solid phase, that is, the neutral fraction of the free DCF concentration. The data presented in Figures 3a and 4a show that the concentration of free diclofenac in the sample decreases significantly as the concentration of sorbing NPs is increased. On the other hand, the rate of achieving partition equilibrium is governed by the total concentration of free and labile species, together with their mean diffusion coefficient. The normalized temporal accumulation profiles of diclofenac in the presence of SiO2 nanoparticles (Figure 3b) or BSA (Figure 4b) highlight that partition equilibrium is attained faster in the presence of sorbing NPs. The experimental data agree quite well with the steady-state coupled diffusion model, as expressed by a combination of eqs 12 and 14 for labile systems containing NPs and considering the free neutral diclofenac to be the only accumulated species. This result unambiguously demonstrates that the NP species themselves are not significantly partitioning between the solid phase and the aqueous medium. Still, they do contribute to the diffusive transport of DCF: because the DCF/

Figure 2. Characteristics of the adsorption of diclofenac by SiO2 nanoparticles at pH 5.0, I = 0.001 M. The points (◆) are the experimental data, and the dashed curve is the fit to the Henry isotherm (eq 1). Total concentration of particles in the dispersion = 8.2 × 10−3 mol m−3.

between diclofenac and BSA is proposed to involve binding of the negatively charged diclofenac species to positively charged amino acid residues in the protein,41 although BSA carries a modest net negative charge at pH 7.4.44 Predicted Lability of Diclofenac−NP complexes. The rate constant for association of diclofenac with the NPs was computed from eq 3, that is, for the usual case in which the overall rate of adsorption is determined by the rate of diffusive supply of DCF to the surface of the spherical NP. This yields kad values of 2.37 × 107 m3 mol−1 s−1 for silica and 1.23 × 107 m3 mol−1 s−1 for BSA. In combination with the thermodynamic affinity constants, the derived rate constants for desorption are 2.3 × 106 and 1.2 × 105 s−1 for silica and BSA, respectively. Such high values of kdes render these systems dynamic on the level of the bulk dispersion on all time scales of interest in the present study (eq 5).45 For the highest extent of binding measured, the lability criterion (eq 11) is computed to be ca. 4 × 103 for the hard silica NP and ca. 5 for the permeable BSA NP; that is, both complex systems may be expected to be labile in the process of transfer of diclofenac molecules through the macroscopic SPME surface. Temporal Accumulation of Diclofenac in SPME fibers. Under our experimental conditions, diclofenac is present in both deprotonated (mononegatively charged) and neutral forms (log K1H = 3.95 (I = 0.05 M),46 4.21 (I = 0.025 M)47). The experimental data show that the concentration of diclofenac accumulated in solid phase decreases dramatically with increasing pH, in agreement with previous SPME reports on charged/neutral forms of organics.48−50 Furthermore, the change in partition of diclofenac with increasing pH corresponds well with the change in the fraction of protonated diclofenac in the sample solution (Table 1). It can therefore be assumed that essentially only the neutral protonated species partitions into the solid phase. Thus, the true thermodynamic solid/water partition coefficient, Ksw, as computed from the concentration of neutral protonated diclofenac in the bulk solution, is found to be approximately independent of pH (Table 1). The extent of ionization of both diclofenac and the SiO2 nanoparticles is dependent on pH and ionic strength, and SiO2 14676

dx.doi.org/10.1021/la303143w | Langmuir 2012, 28, 14672−14680

Langmuir

Article

Figure 3. Mean concentration of diclofenac in the PA solid-phase film, cs̅ ,DCF, as a function of extraction time, t, in the absence and presence of different concentrations of SiO2 nanoparticles at pH 5: (a) absolute form and (b) normalized with respect to the eventual equilibrium * 0 (only highest degree of sorption shown for value, ceq s,DCF = KSWcw,DCF clarity). Points are experimental data; dashed curves are computed from eqs 12 and 14, with the pertaining concentration of the accumulated form of free diclofenac, c*w,DCF0, for the given total concentration of DCF, 1.57 × 10−2 mol m−3.

Figure 4. Mean concentration of diclofenac in the PEG solid-phase film, cs̅ ,DCF, as a function of extraction time, t, in the absence and presence of BSA at pH 7.4: (a) absolute form and (b) normalized with respect to the eventual equilibrium value, ceq s,DCF = KSWc* w,DCF0. Points are experimental data; dashed curves are computed from eqs 12 and 14, and the solid line is the best fit to the experimental data in the presence of BSA (k̅X = 1.4 × 10−5 s−1). Computations were performed with the pertaining concentration of the accumulated form of free * 0, for the given total concentration of DCF, 8 × 10−2 diclofenac, cw,DCF mol m−3.

NP complex is labile, it acts as a shuttle that carries DCF molecules from the bulk medium to the solid-phase surface. The experimentally measured accumulation rate constants (calculated from plots of ln(1 − cs̅ ,X/c*w,X0Ksw) versus time, given in the Supporting Information) and those computed via eq 14 are given in Tables 2 and 3 for SiO2 and BSA, respectively. The convincing agreement between the experimental and theoretical accumulation rate constants confirms the dynamic equilibrium between the bound and free diclofenac species in the bulk medium,8,55 and labile behavior of particle-bound diclofenac on the effective time scale of diffusion toward the solid phase. Consequently, the rate-limiting step of the extraction process is determined by the coupled diffusion of the free and bound forms of the analyte in the aqueous phase. Similar conclusions were reached for atrazine in the presence of sorbing latex NPs.8

Table 2. Rate Constant, kDCF, for SPME Accumulation of Diclofenac from Aqueous Dispersions of 7.5 nm SiO2 Nanoparticles at pH 5 SiO2 NP conc/ 10−3 mol m−3

% diclofenac bound

δ̅/ μma

D̅ /10−10 m2 s−1b

kDCF,exp/ 10−5 s−1c

k̅DCF,theor/ 10−5 s−1d

0 4.1 8.2 16.5

0 34 48 65

54 48 44 40

3.90 2.67 2.16 1.54

2.5 3.4 5.0 6.1

3.5 4.6 5.8

a δ for the diclofenac-only case is determined from experiment (eqs 12 and 13), and δ̅ values for the dispersions are then derived assuming laminar convection.53 bCalculated via eq 10, using D for silica NPs equal to (2.7 ± 0.3) × 10−11 m2 s−1.54 cDetermined by fitting eq 12 to the experimental SPME accumulation curves (Figure 3). dCalculated via eq 14, As = 2.1 × 10−5 m2, Vs = 5.2 × 10−10 m3, Ksw = 8.3 × 103.



CONCLUSION A conceptual framework is presented for describing the chemodynamics of NP-bound organic compounds under conditions of their speciation analysis by SPME. For given

spatial scale and time scale, the lability of organic−NP complexes can be computed from their thermodynamic affinity constants and rates of adsorption and/or desorption (eq 11). If 14677

dx.doi.org/10.1021/la303143w | Langmuir 2012, 28, 14672−14680

Langmuir



Table 3. Rate Constant, kDCF, for SPME Accumulation of Diclofenac from an Aqueous Dispersion of BSA Nanoparticles at pH 7.4

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Notes

BSA conc/10−2 mol m−3

% diclofenac bound

δ̅/ μma

D̅ /10−10 m2 s−1b

kDCF,exp/ 10−5 s−1c

k̅DCF,theor/ 10−5 s−1d

0 8.0

0 93

44 26

3.90 0.85

1.4 6.6

7.3

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.

a δ for the diclofenac-only case is determined from experiment (eqs 12 and 13), and δ̅ values for the dispersions are then derived assuming laminar convection.53 bCalculated via eq 10, using D for BSA equal to 6.2 × 10−11 m2 s−1 (pH 7.4, 0.15 M NaCl).33 cDetermined by fitting eq 12 to the experimental SPME accumulation curves (Figure 4). d Calculated via eq 14, As = 8 × 10−6 m2, Vs = 3.6 × 10−10 m3, Ksw = 1.44 × 104.



SYMBOLS AND ABBREVIATIONS BSA = bovine serum albumin DCF = diclofenac NP = nanoparticle SPME = solid-phase microextraction X = target organic compound ANP = total surface area of the particles, m2 As = surface area of the solid phase, m2 cS = smeared-out volume concentration of binding sites, mol m−3 c*w,X = bulk aqueous concentration of free X (all protonated forms), mol m−3 c*w,X0 = bulk aqueous concentration of the free neutral form of X, mol m−3 ceq s,X = bulk concentration of X in the solid phase at partition equilibrium (=Kswc*w,X0), mol m−3 cS,max = maximum number of binding sites for permeable NPs, mol m−3 cX−S = local volume concentration of X within the particle, mol m−3 cX̅ −S = smeared-out volume concentration of NP-bound X, mol m−3 cs̅ ,X = mean concentration of X in the solid phase, mol m−3 c0s,X = concentration of X at the water−solid interphase, mol m−3 ct,X * = total bulk concentration of all X species in the aqueous sample, mol m−3 c*lab,X = concentration of labile X species in the medium, mol m−3 ds = thickness of the solid-phase coating, m Dw,X = diffusion coefficient of X in water, m2 s−1 Ds,X = diffusion coefficient of X in the solid phase, m2 s−1 D̅ = mean diffusion coefficient, m2 s−1 DX−S = diffusion coefficient of the adsorbed X, m2 s−1 J = analyte flux from sample medium to solid phase, mol m−2 s−1 Jkin = desorption rate limited flux, mol m−2 s−1 Jdif * = limiting diffusive flux, mol m−2 s−1 kX = accumulation rate constant, s−1 k̅X = effective accumulation rate constant for the complex system, s−1 kad = overall rate constant for adsorption, m3 mol−1 s−1 k′ad = the product of kad and the concentration of sites, kadcS, s−1 kdif = rate constant for diffusive supply of X to the spherical NP surface, m3 mol−1 s−1 kdes = overall rate constant for desorption, s−1 KX = thermodynamic adsorption constant, m3 mol−1 Ksw = solid phase/water partition coefficient 3 = lability criterion parameter

the rate constants are not known, then kad for neutral organic targets can be estimated from their rate of diffusive transport to the NP surface (eq 3). We highlight that measurement of the time dependence of accumulation in the solid phase is essential to identify the presence of labile target organic species. Determination of only the extent of equilibrium partitioning merely identifies the concentration of unbound organic species, and often only the neutral form thereof. Complex organic species that are labile on the time scale of SPME accumulation are potentially able to enhance the supply flux of free organic molecules to other accumulating surfaces, for which the time scale of the pertaining process is comparable to or greater than that of the SPME extraction, for example, organisms in the size range of micrometer or gill lamellae. The solid phase in SPME is typically chosen to have a high Ksw for the target compound; however, there is a paucity of equivalent partitioning constants applicable to organisms.56 It has been shown that diffusive mass transport in the aqueous medium is rate limiting for uptake of hydrophobic chemicals by fish.57,58 Analogous with concepts applicable to metal complex species,59 the lability of organic complexes is expected to be independent of the boundary conditions at the accumulating surface, because the experiment typically features a Ksw ≫ 1 (so that diffusion in the medium will be accumulation rate-limiting). Therefore, the effective time scale of accumulation is the pertinent factor in determining whether an organic complex species can dissociate and contribute to the supply flux of the free form. It thus appears that SPME provides a tool to characterize the dynamic features of organic chemicals in the presence of sorbing NPs. However, in natural systems, the behavior of trace organics such as pharmaceuticals could be involved: metabolites and transformation products60 may have sorption behavior different from that of the parent compound and may bind to a variety of sorbents in aquatic systems.61,62 Each associate will exhibit its own particular kinetic features, and thus a differentiated approach may be required to characterize the chemodynamic features and correctly quantify the fate and reactivity of organic targets in natural waters.



Article

ASSOCIATED CONTENT

S Supporting Information *

Logarithmic representation of the temporal accumulation of diclofenac in an SPME solid phase in the presence of different concentrations of SiO2 nanoparticles. This material is available free of charge via the Internet at http://pubs.acs.org. 14678

dx.doi.org/10.1021/la303143w | Langmuir 2012, 28, 14672−14680

Langmuir

Article

(15) Cui, X. Y.; Jia, F.; Chen, Y. X.; Gan, J. Influence of single-walled carbon nanotubes on microbial availability of phenanthrene in sediment. Ecotoxicology 2011, 20, 1277−1285. (16) van Leeuwen, H. P.; Buffle, J. Chemodynamics of aquatic metal complexes: from small ligands to colloids. Environ. Sci. Technol. 2009, 43, 7175−7183. (17) van Leeuwen, H. P.; Town, R. M.; Buffle, J.; Cleven, R. F. M. J.; Davison, W.; Puy, J.; van Riemsdijk, W. H.; Sigg, L. Dynamic speciation analysis and bioavailability of metals in aquatic systems. Environ. Sci. Technol. 2005, 39, 8545−8556. (18) Gdaniec-Pietryka, M.; Wolska, L.; Namieśnik, J. Physical speciation of polychlorinated biphenyls in the aquatic environment. Trends Anal. Chem. 2007, 26, 1005−1012. (19) ter Laak, T. L.; ter Bekke, M. A.; Hermens, J. L. M. Dissolved organic matter enhances transport of PAHs to aquatic organisms. Environ. Sci. Technol. 2009, 43, 7212−7217. (20) Lindqvist, N.; Tuhkanen, T.; Kronberg, L. Occurrence of acidic pharmaceuticals in raw and treated sewages and in receiving waters. Water Res. 2005, 39, 2219−2228. (21) Ternes, T. A. Occurrence of drugs in German sewage treatment plants and rivers. Water Res. 1998, 32, 3245−3260. (22) Morel, F. M. M.; Hering, J. G. Principles and Applications of Aquatic Chemistry; Wiley: New York, 1993. (23) van Leeuwen, H. P.; Town, R. M.; Buffle, J. Impact of ligand protonation on Eigen-type metal complexation kinetics in aqueous systems. J. Phys. Chem. A 2007, 111, 2115−2121. (24) van Leeuwen, H. P.; Town, R. M.; Buffle, J. Chemodynamics of soft nanoparticulate metal complexes in aqueous media: basic theory for spherical particles with homogeneous spatial distributions of sites and charges. Langmuir 2011, 27, 4514−4519. (25) van Leeuwen, H. P.; Buffle, J.; Town, R. M. Electric relaxation processes in chemodynamics of aqueous metal complexes: from simple ligands to soft nanoparticulate complexants. Langmuir 2012, 28, 227− 234. (26) Louch, D.; Motlagh, S.; Pawliszyn, J. Dynamics of organic compound extraction from water using liquid-coated fused silica fibers. Anal. Chem. 1992, 64, 1187−1199. (27) van Leeuwen, H. P.; Cleven, R.; Buffle, J. Voltammetric techniques for complexation measurements in natural aquatic media. Role of the size of macromolecular ligands and dissociation kinetics of complexes. Pure Appl. Chem. 1989, 61, 255−274. (28) Pygall, S. R.; Griffiths, P. C.; Wolf, B.; Timmins, P.; Melia, C. D. Solution interactions of diclofenac sodium and meclofenamic acid sodium with hydroxypropyl methyl cellulose. Int. J. Pharm. 2011, 405, 55−62. (29) Heyrovský, J.; Kuta, J. Principles of Polarography; Academic Press: New York, 1966. (30) Heringa, M. B.; Hermens, J. L. M. Measurement of free concentrations using negligible depletion-solid phase microextraction (nd-SPME). Trends Anal. Chem. 2003, 22, 575−587. (31) Heringa, M. B.; Hogevonder, C.; Busser, F.; Hermens, J. L. M. Measurement of the free concentration of octylphenol in biological samples with negligible depletion-solid phase microextraction (ndSPME): Analysis of matrix effects. J. Chromatogr., B 2006, 834, 35−41. (32) Benhabib, K.; ter Laak, T. L.; van Leeuwen, H. P. Steady-state diffusion regime in solid-phase microextraction kinetics. Anal. Chim. Acta 2008, 609, 113−119. (33) Meechai, N.; Jamieson, A. M.; Blackwell, J. Translational diffusion coefficients of bovine serum albumin in aqueous solution at high ionic strength. J. Colloid Interface Sci. 1999, 218, 167−175. (34) Fini, A.; Fazio, G.; Gonzalez-Rodriguez, M.; Cavallari, C.; Passerini, N.; Rodriguez, L. Formation of ion-pairs in aqueous solutions of diclofenac salts. Int. J. Pharm. 1999, 187, 163−173. (35) Maitani, Y.; Nakagaki, M.; Nagai, T. Determination of the acid dissociation constants in ethanol-water mixtures and partition coefficients for diclofenac. Int. J. Pharm. 1991, 74, 105−116. (36) Barbato, F.; Caliendo, G.; La Rotonda, M.; Silipo, C.; Toraldo, G.; Vittoria, A. Distribution coefficients by curve fitting: application to

NNP = total number of nanoparticles per unit volume of dispersion Rad = overall adsorption rate, mol m−3 s−1 Rdes = overall desorption rate, mol m−2 s−1 rp = particle radius, m t = time variable V = volume of the dispersion, m3 VNP = volume of a nanoparticle, m3 Vs = volume of the accumulating solid phase, m3 ΓX−S = surface concentration of X bound to impermeable NPs, mol m−2 ΓS,max = maximum number of binding sites for impermeable NPs, mol m−2 δ = thickness of the diffusion layer, m δ̅ = mean thickness of the steady-state diffusion layer, m μ = reaction layer thickness, m τss = characteristic time to establishing steady-state diffusion, s



REFERENCES

(1) Perazzolo, C.; Morasch, B.; Kohn, T.; Magnet, A.; Thonney, D.; Chèvre, N. Occurrence and fate of micropollutants in the Vidy Bay of Lake Geneva, Switzerland. Part I: Priority list for environmental risk assessment of pharmaceuticals. Environ. Toxicol. Chem. 2010, 29, 1649−1657. (2) Heberer, T. Occurrence, fate, and removal of pharmaceutical residues in the aquatic environment: a review of recent research data. Toxicol. Lett. 2002, 131, 5−17. (3) Brooks, B. W.; Huggett, D. B.; Boxall, A. B. A. Pharmaceuticals and personal care products: research needs for the next decade. Environ. Sci. Technol. 2009, 28, 2469−2472. (4) la Farré, M.; Pérez, S.; Kantiani, L. Fate and toxicity of emerging pollutants, their metabolites and transformation products in the aquatic environment. Trends Anal. Chem. 2008, 27, 991−1007. (5) Belardi, R. P.; Pawliszyn, J. B. The application of chemically modified fused silica fibers in the extraction of organics from water matrix samples and their rapid transfer to capillary columns. Water Pollut. Res. J. Can. 1989, 24, 179−191. (6) Wu, J.; Mester, Z.; Pawliszyn, J. Speciation of organoarsenic compounds by polypyrrole-coated capillary in-tube solid phase microextraction couple with liquid chromatography/electrospray ionization mass spectrometry. Anal. Chim. Acta 2000, 424, 211−222. (7) Mester, Z.; Sturgeon, R.; Pawliszyn, J. Solid phase microextraction as a tool for trace element speciation. Spectrochim. Acta, Part B 2001, 56, 233−260. (8) Benhabib, K.; Town, R. M.; van Leeuwen, H. P. Dynamic speciation analysis of atrazine in aqueous latex nanoparticle dispersions using solid phase microextraction (SPME). Langmuir 2009, 25, 3381− 3386. (9) Nowack, B.; Bucheli, T. D. Occurrence, behavior and effects of nanoparticles in the environment. Environ. Pollut. 2007, 150, 5−22. (10) Buffle, J., van Leeuwen, H. P., Eds. Environmental Particles; Lewis Publishers: MI, 1992; Vol. I, Chapters 10−13. (11) Hofmann, T.; von der Kammer, F. Estimating the relevance of engineered carbonaceous nanoparticle facilitated transport of hydrophobic organic contaminants in porous media. Environ. Pollut. 2008, 157, 1117−1126. (12) Tungittiplakorn, W.; Cohen, C.; Lion, L. W. Engineered polymeric nanoparticles for bioremediation of hydrophobic contaminants. Environ. Sci. Technol. 2005, 39, 1354−1358. (13) Yang, Y.; Shu, L.; Wang, X.; Xing, B.; Tao, S. Mechanisms regulating bioavailability of phenanthrene sorbed on a peat soil-origin humic substance. Environ. Toxicol. Chem. 2012, 31, 1431−1437. (14) Farkas, J.; Nizzetto, L.; Thomas, K. V. The binding of phenanthrene to engineered silver and gold nanoparticles. Sci. Total Environ. 2012, 425, 283−288. 14679

dx.doi.org/10.1021/la303143w | Langmuir 2012, 28, 14672−14680

Langmuir

Article

ionogenic nonsteroidal anti-inflammatory drugs. Struct.-Act. Relat. 1986, 5, 88−95. (37) Leng, X.; Starchev, K.; Buffle, J. Adsorption of fluorescent dyes on oxide nanoparticles studied by fluorescence correlation spectroscopy. Langmuir 2002, 18, 7602−7608. (38) Sudlow, G.; Birkett, D. J.; Wade, D. N. The characterization of two specific drug binding sites on human serum albumin. Mol. Pharmacol. 1975, 11, 824−832. (39) Sudlow, G.; Birkett, D. J.; Wade, D. N. Further characterization of specific drug binding sites on human serum albumin. Mol. Pharmacol. 1976, 12, 1052−1061. (40) Alarcón, E.; Aspée, A.; Abuin, E. B.; Lissi, E. A. Evaluation of solute binding to proteins and intra-protein distances from steady state fluorescence measurements. J. Photochem. Photobiol., B 2012, 106, 1− 17. (41) Sharma, R.; Choudhary, S.; Kishore, N. Insights into the binding of the drugs diclofenac sodium and cefotaxime sodium to serum albumin: calorimetry and spectroscopy. Eur. J. Pharm. Sci. 2012, 46, 435−445. (42) Hosseinzadeh, R.; Maleki, R.; Matin, A. A. Interaction of diclofenac with bovine serum albumin investigated by dicloflenacselective electrode. Acta Chim. Slov. 2007, 54, 126−130. (43) Wang, D.; Zhang, Y.; Liu, Y.-N.; Wang, J. Estimation of binding constants for diclofenac sodium and bovine serum albumin by affinity capillary electrophoresis and fluorescence spectroscopy. J. Liq. Chromatogr. Relat. Technol. 2008, 31, 2077−2088. (44) Salis, A.; Boström, M.; Medda, L.; Cugia, F.; Barse, B.; Parsons, D. F.; Ninham, B. W.; Monduzzi, M. Measurements and theoretical interpretation of points of zero charge/potential of BSA protein. Langmuir 2011, 27, 11597−11604. (45) van Leeuwen, H. P. Speciation dynamics and bioavailability of metals. J. Radioanal. Nucl. Chem. 2000, 236, 487−492. (46) Zhou, C.; Jin, Y.; Kenseth, J. R.; Stella, M.; Wehmeyer, K. R.; Heineman, W. R. Rapid pKa estimation using vacuum-assisted multiplexed capillary electrophoresis (VAMCE) with ultraviolet detection. J. Pharm. Sci. 2005, 94, 576−589. (47) Wan, H.; Holmén, A. G.; Wang, Y.; Lindberg, W.; Englund, M.; Någård, M. B.; Thompson, R. A. High-throughput screening of pKa values of pharmaceuticals by pressure-assisted capillary electrophoresis and mass spectrometry. Rapid Commun. Mass Spectrom. 2003, 17, 2639−2648. (48) Lord, H.; Pawliszyn, J. Microextraction of drugs. J. Chromatogr., A 2000, 902, 17−63. (49) Rodríguez, I.; Carpinteiro, J.; Quintana, J. B.; Carro, A. M.; Lorenzo, R. A.; Cela, R. Solid-phase microextraction with on-fiber derivatization for the analysis of anti-inflammatory drugs in water samples. J. Chromatogr., A 2004, 1024, 1−8. (50) Sarafraz-Yazdi, A.; Amiri, A.; Rounaghi, G.; Eshtiagh-Hosseini, H. Determination of non-steroidal anti-inflammatory drugs in water samples by solid-phase microextraction based sol-gel technique using poly(ethylene glycol) grafted multi-walled carbon nanotubes coated fiber. Anal. Chim. Acta 2012, 720, 134−141. (51) Bui, T. H.; Choi, H. Adsorptive removal of selected pharmaceuticals by mesoporous silica SBA-15. J. Hazard. Mater. 2009, 168, 602−608. (52) Drillia, P.; Stamatelatou, K.; Lyberatos, G. Fate and mobility of pharmaceuticals in solid matrices. Chemosphere 2005, 60, 1034−1044. (53) Levich, V. G. Physicochemical Hydrodynamics; Prentice-Hall: Englewood Cliffs, NJ, 1962. (54) Goveia, D.; Pinheiro, J. P.; Milkova, V.; Rosa, A. H.; van Leeuwen, H. P. Dynamics and heterogeneity of Pb(II) binding by SiO2 nanoparticles in an aqueous dispersion. Langmuir 2011, 27, 7877− 7883. (55) Kramer, N. I.; van Eijkeren, J. C. H.; Hermens, J. L. M. Influence of albumin on sorption kinetics in solid-phase microextraction: consequences for chemical analyses and uptake processes. Anal. Chem. 2007, 79, 6941−6948.

(56) 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. (57) Sijm, D. T. H. M.; van der Linde, A. Size-dependent bioconcentration kinetics of hydrophobic organic chemicals in fish based on diffusive mass transfer and allometric relationships. Environ. Sci. Technol. 1995, 29, 2769−2777. (58) Gobas, F. A. P. C.; Mackay, D. Dynamics of hydrophobic organic chemical bioconcentration in fish. Environ. Toxicol. Chem. 1987, 6, 495−504. (59) Zhang, Z.; Alemani, D.; Buffle, J.; Town, R. M.; Wilkinson, K. J. Metal flux through consuming interfaces in ligand mixtures: boundary conditions do not influence the lability and relative contributions of metal species. Phys. Chem. Chem. Phys. 2011, 13, 17606−17614. (60) Canonica, S.; Meunier, L.; von Gunten, U. Phototransformation of selected pharmaceuticals during UV treatment of drinking water. Water Res. 2008, 42, 121−128. (61) Margon, A.; Pastrello, A.; Mosetti, D.; Cantone, P.; Leita, L. Interaction between diclofenac and soil humic acids. Soil Sediment Contam. 2009, 18, 489−496. (62) Yang, K.; Xing, B. Sorption of phenanthrene by humic acidcoated nanosized TiO2 and ZnO. Environ. Sci. Technol. 2009, 43, 1845−1851.

14680

dx.doi.org/10.1021/la303143w | Langmuir 2012, 28, 14672−14680