from a natural sediment

Science of the Total Environment 472 (2014) 273–281

Contents lists available at ScienceDirect

Science of the Total Environment journal homepage: www.elsevier.com/locate/scitotenv

Sorption/desorption of non-hydrophobic and ionisable pharmaceutical and personal care products from reclaimed water onto/from a natural sediment Virtudes Martínez-Hernández a,⁎, Raffaella Meffe a, Sonia Herrera a, Elena Arranz b, Irene de Bustamante a,b a b

IMDEA Agua, Madrid Institute for Advanced Studies in Water, Parque Científico Tecnológico de la Universidad de Alcalá, 28805 Alcalá de Henares, Madrid, Spain University of Alcalá, Geography and Geology Department, 28871 Alcalá de Henares, Madrid, Spain

H I G H L I G H T S • • • • •

Sorption–desorption of wide usage groups of PPCPs to sediment was quantified. Positively ionized PPCPs showed higher sorption than negatively and neutral ones. More than 70% of the total sorption was due to interaction with mineral surfaces. No competition of PPCPs with inorganic ions was observed in desorption processes. The potential of PPCPs to infiltrate decreases as follows CBZ N ACP N NPX N ATN N SXZ N CAF.

a r t i c l e

i n f o

Article history: Received 13 September 2013 Received in revised form 5 November 2013 Accepted 5 November 2013 Available online 30 November 2013 Keywords: PPCPs Reclaimed water Sorption–desorption Unsaturated zone Groundwater

a b s t r a c t In the present work, the sorption of pharmaceutical and personal care products (PPCPs) (acetaminophen, atenolol, carbamazepine, caffeine, naproxen and sulphamethoxazole) onto the natural organic matter (NOM) and the inorganic surfaces of a natural sandy loam sediment was quantified separately. The quantification was based on the PPCP charge, their degree of ionisation, their octanol–water partitioning coefficient (KOW) and the sediment organic carbon fraction (ƒOC). PPCP desorption from the sediment was examined under conditions of infiltrating water containing a high concentration of inorganic ions (mimicking infiltrating reclaimed water), and a low concentration (and smaller diversity) of inorganic ions (mimicking rainwater infiltration). Batch tests were performed using a sediment/water ratio of 1:4 and a PPCP initial concentration ranging from 1 to 100 μg L−1. The results showed the type and degree of PPCP ionisation to strongly influence the sorption of these compounds onto the sediment. The sorption of cationic species onto the sediment was higher than that of anionic species and mostly reversible; the sorption of neutral species was negligible. The anionic species sorbed less onto the sediment, but also desorbed less easily. More than 70% of the total sorption was due to interaction with mineral surfaces. This holds especially true for cationic species (atenolol and caffeine) which sorption was enhanced by the negative surface charge of the sediment. The presence of inorganic ions had no impact on the desorption of the PPCPs from the sediment. According to the calculated percentages of removal, the mobility followed the order: carbamazepine N acetaminophen N naproxen N atenolol N sulfamethoxazole N caffeine. © 2013 Elsevier B.V. All rights reserved.

1. Introduction The problems of an increasing population, water resource contamination and climate change have led to the scarcity of water in many countries, especially in the Middle East and Mediterranean (Lazarova et al., 2001). The use of reclaimed water may often provide a technically and economically feasible solution (Kiziloglu et al., 2008; Molinos-Senante et al., 2011), but its use in irrigation and/or aquifer recharge can introduce a range of contaminants into the ⁎ Corresponding author. Tel.: +34 918305962; fax: +34 918305961. E-mail address: [email protected] (V. Martínez-Hernández). 0048-9697/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.scitotenv.2013.11.036

environment if these are not removed during wastewater treatment (Levine and Asano, 2004; Toze, 2006). Indeed, a number of pharmaceutical and personal care products (PPCPs) are very often found in different water compartments. Several studies have reported PPCPs present in Spanish surface water and groundwater (Estévez et al., 2012; García-Galán et al., 2010; González Alonso et al., 2010; Jurado et al., 2012; Koeck-Schulmeyer et al., 2011; Martinez Bueno et al., 2010; Teijon et al., 2010). In general, groundwater PPCP concentrations are lower than those of surface and wastewater (ng L− 1 compared to μg L− 1), suggesting that these compounds are subject to retention and/or degradation during infiltration (Jurado et al., 2012).

274

V. Martínez-Hernández et al. / Science of the Total Environment 472 (2014) 273–281

The mobility of PPCPs in the environment depends essentially on their sorption onto and desorption from sediments, and their degradation (physical and biological) during infiltration (Sabljic et al., 1995). These processes clearly affect their bioavailability (Katayama et al., 2010). In the last 10 years, several batch experiments investigating sorption of PPCPs onto sediments have been carried out (Chefetz et al., 2008; Durán-Álvarez et al., 2012; Fenet et al., 2012; Karnjanapiboonwong et al., 2010; Kibbey et al., 2007; Lin et al., 2010; Lorphensri et al., 2007; Ramil et al., 2009; Stein et al., 2008; Xu et al., 2009; Yamamoto et al., 2009; Yu et al., 2013). Some of them pointed out a direct correlation between the amount of sediment organic matter and the degree of sorption (Chefetz et al., 2008; Fenet et al., 2012; Kibbey et al., 2007; Yu et al., 2013). However, other studies have also shown that sediment inorganic surfaces may affect PPCPs sorption (Lin et al., 2010; Pan et al., 2009; Schaffer et al., 2012a; Tolls, 2001; Yamamoto et al., 2009) and therefore this process should be taken into account besides partitioning onto natural organic matter (NOM). Only recently, the degree of ionisation of some functional groups of PPCPs at environmental pH has been recognized as an additional factor that may control the sorption of these compounds onto natural sediments (Schaffer et al., 2012b). Only few articles have included desorption essays in their research (Chefetz et al., 2008; Durán-Álvarez et al., 2012; Karnjanapiboonwong et al., 2010; Lin et al., 2010; Stein et al., 2008; Teijón et al., 2013) and results showed cases of hysteretic and reversible sorption of several PPCPs. This emphasizes the importance of linking/coupling sorption and desorption studies to correctly assess the mobility of PPCPs through the subsurface. Information on the sorption/desorption behaviour of PPCPs in sediments is, however, conflicting when it is available. For example, Lin and Gan (2011) reported moderate to strong sorption onto soil for an anti-inflammatory drug and Durán-Álvarez et al. (2012) found low sorption rates for the same compound in two soils. Ramil et al. (2009) and Yamamoto et al. (2009) identified different sorption degrees onto sediments for a drug belonging to the group of β-blockers. Drillia et al. (2005) showed a higher degree of sorption for a sulphonamide antibiotic than that described by Stein et al. (2008) for the same contaminant. Furthermore, papers in this area commonly fail to separate the individual effects of PPCP ionisation and partitioning (Durán-Álvarez et al., 2012; Karnjanapiboonwong et al., 2010; Löffler et al., 2005; Lorphensri et al., 2007; Stein et al., 2008; Williams et al., 2009). Studies on how sorption/desorption onto/from sediments affects their transport are therefore needed. Competition between inorganic ions and PPCPs also occurs during the latters' sorption onto sediments and has only recently been suspected to influence PPCP concentrations in infiltrated water (Niedbala et al., 2013; Schaffer et al., 2012a). Once sorbed onto sediments they may be desorbed again by water carrying dissolved ions, e.g., reclaimed irrigation water or by water carrying few inorganic ions, e.g., rainwater. The present paper focuses on a number of compounds with different chemical properties representative of a range of PPCP classes: the analgesic acetaminophen (ACP), the β-blocker atenolol (ATN), the anticonvulsant carbamazepine (CBZ), the stimulant caffeine (CAF), the non-steroidal anti-inflammatory drug naproxen (NPX), and the antibiotic sulfamethoxazole (SXZ). All these compounds are nonhydrophobic chemicals. They can be either neutral or positively or negatively ionised depending on the pH of the environment. With the exception of ACP, all have been detected at concentrations of N 100 ng L − 1 in Spanish groundwater (Jurado et al., 2012). These high concentrations, together with the limited information available on their sorption/desorption onto/from sediments point to a special concern for their environmental fate. The aims of this study are: i) to investigate the impact of PPCP ionization (neutral, positive, negative) on the extent of sorption– desorption processes; ii) to separately quantify PPCP sorption onto

the natural organic matter and the inorganic surfaces and discuss potential sorption mechanisms; iii) to evaluate the possibility of competition between PPCPs and inorganic ions through a desorption test that contemplates two scenarios: the infiltration of reclaimed irrigation water carrying dissolved inorganic ions, and the infiltration of rainwater carrying few ions. 2. Materials and methods 2.1. Sediment sampling and analysis Sediment samples were collected from the unsaturated zone of the Manzanares-Jarama groundwater body, part of the Madrid Detrital Tertiary Aquifer. From a hydrogeological point of view, this area can be divided into 3 sectors with similar characteristics (Torres et al., 1995). The collected samples belong to the middle sector, characterized by loamy and sandy sediments, that constitutes the recharge area of the main aquifer. Moreover, to avoid misleading of experimental results, sampling was carried out in an area free of PPCP contamination sources. Sediment samples were air-dried, gently crushed and passed through a 2 mm sieve. pH and electrical conductivity (EC) were measured in a sediment–water suspension (sediment–water ratio 1:2.5 and 1:5 respectively). Particle size distribution was determined following the method of Gee and Bauder (1986). Part of the air-dried, sieved sample was crushed for the determination of the sediment organic carbon fraction (ƒOC) following the method of Nelson and Sommers (1982). Oxalate-extractable Fetot and Altot were obtained by a 4 h extraction of 1 g of dried sediment in 100 mL of 0.2 M acid ammonium oxalate at pH 3 (McKeague and Day, 1966). The cation exchange capacity (CEC) was determined at natural sediment pH by extracting 5 g of dried sediment with 1 M sodium acetate and 1 M ammonium acetate. The potentiometric titration method was used to determine the sediment point of zero charge (PZC). The properties of the sandy loam sediment are summarised in Table 1. 2.2. Synthesis of reclaimed water Samples of reclaimed water (n = 5) were collected from wastewater treatment plant secondary effluent (which fulfilled the requirements for water reuse (BOE, 2007)) from the study area. Standard analyses were performed (see the Section 2.6) to determine the inorganic ion composition and dissolved organic carbon content (DOC) of these samples, and the means recorded. Using this information, a stock solution of synthesized reclaimed water (SRW) was then produced dissolving the following reagents (purity N 95.0%) in tap water: NH4Cl (0.07 g L−1), MgSO4 (0.1 g L − 1 ), CaCl 2 (0.01 g L − 1 ), K 2 HPO 4 (0.02 g L − 1 ), NaHCO 3 (0.25 g L− 1), peptone (0.01 g L− 1), and meat extract (0.01 g L− 1) (all purchased from Scharlab, Spain). Table 2 shows the final composition of the SRW. 2.3. Target PPCPs The target compounds – ACP, ATN, CAF, CBZ, NPX, paraxanthine (PXT) and SXZ (purity N 98%) – were purchased from Sigma-Aldrich. All were dissolved in methanol (99.9%) from Sigma-Aldrich. 2.4. Sorption experimental design The sorption isotherms of the studied PPCPs were determined in parallel batch experiments following OECD guideline 106 (OECD, 2000). SRW (200 mL) in 1 L glass vessels containing 50 g sterilized sediment was spiked with PPCPs at different concentrations (1, 5, 10, 25, 50, 75 and 100 μg L−1). The sediment and vessels were autoclaved for 15 min at 121 ºC. It was believed that the conventional use of 10% sodium azide to inhibit microbial activity might affect the sorption capacity of the sediment (Trevors, 1996). The final methanol

V. Martínez-Hernández et al. / Science of the Total Environment 472 (2014) 273–281

275

Table 1 Sediment properties. pH

PZC

EC (μS cm−1)

Sand (%)

Silt (%)

Clay (%)

ƒOC (%)

Fe oxide (%)

Al oxide (%)

CEC (cmol kg−1)

6.56

6.80

15.10

82.50

10.00

7.50

1.44

0.22

0.69

5.41

concentrations were kept at b0.1% (v/v) to avoid co-solvent effects. The OECD guidelines (OECD, 2000) recommend a 1:5 sediment/ water ratio for batch experiments, but a 1:4 sediment/SRW ratio was selected to better mimic non-saturated water conditions. The vessels were shaken at 140 rpm for 24 h (as determined in preliminary experiments) until sorption equilibrium was reached. All preparations were made in sextuplicate. To confirm that the original sediment was PPCP-free, and to exclude the possibility of PPCP-sorption onto the vessels and degradation, control (without sediment) and blank (without PPCPs) samples were prepared in triplicate and analysed along with the others. All vessels were wrapped in aluminium foil to minimize photochemical reactions. Oxic conditions were maintained during all experiments. After 24 h, samples were collected from each vessel. To separate the sediment and liquid phases, samples were centrifuged at 5000 rpm for 20 min. To remove the suspended solids and particulate matter, samples were filtered through 0.45 μm membranes and the pH adjusted to 3. To avoid sorption of PPCPs onto filters, teflon membranes were used. These samples were then stored in a freezer at − 18 °C until analysis (see the section 2.6). 2.5. Desorption experimental design

using a two-channel advanced compact ion chromatograph apparatus (Metrohm, Switzerland). The concentrations of PPCPs in the liquid phase were analysed by the direct injection of samples (50 μL) into an LC-MS TripleTOF 5600 system (AB SCIEX, Concord, ON) connected to an HPLC system with an electrospray interface (ESI). The method detection (MDL) and quantification limits (MQL) are 0.02 and 0.07 μg L−1 for CBZ, 0.25 and 0.85 μg L− 1 for ACP, 0.01 and 0.03 μg L− 1 for ATN, 0.10 and 0.30 μg L− 1 for CAF, 0.60 and 2.00 μg L− 1 for NPX, 0.03 and 0.11 μg L−1 for SXZ, 0.20 and 0.65 μg L−1 for PXT. The sorbed concentrations were then calculated by subtraction. Potential CAF degradation was monitored by determining its metabolite PXT. Prior to analysis, all samples were spiked with the mixture of surrogate standards (terbutylazine-D5 and caffeine-C13) in acetonitrile (Sigma-Aldrich). 2.7. Sorption and desorption isotherms The sorption and desorption results for the PPCPs were matched against the linear, Freundlich, and Langmuir isotherm models (Schwarzenbach et al., 2003). The Freundlich model is described as follows: n

Cs ¼ K F  Cw

ð1Þ

The desorption isotherms were determined following OECD guideline 106 as above (OECD, 2000), placing sediment (50 g) with previously sorbed PPCPs (at all concentrations) in contact with either SRW (200 mL) or 0.01 M CaCl2 solution for 24 h in 1 L glass vessels. The use of SRW was intended to simulate the desorption of PPCPs under conditions of infiltration of reclaimed water, while the use of CaCl2 was intended to mimic the infiltration of rainwater. Analysis of the liquid phase was then performed as described below.

where C s (μg kg − 1 ) and C w (μg L − 1 ) are the sorbed and solution concentrations at equilibrium respectively, K F (μg1 − n L n kg − 1 ) is the Freundlich distribution coefficient, and n (dimensionless) is the Freundlich exponent. The linear model is manifested when the Freundlich exponent is equal to 1. Unlike the Freundlich model, the Langmuir contemplates a limited number of sorption sites that become saturated in a monolayer sorbent. It is described as follows:

2.6. Chemical analysis

Cs ¼

Liquid phase (SRW and CaCL2) samples from the sorption and desorption experiments were analysed for their EC, dissolved oxygen − −, 3− (DO) content, DOC, and major dissolved ions (NO− 2 , NO3 PO4 , Cl , + + 2+ 2+ + SO2− , NH Na , K , Ca , Mg ) according to standard methods 4 4 (Eaton et al., 2005). DOC analysis was performed using a total organic carbon analyser (Shimadzu, Japan). The dissolved ions were analysed

Table 2 Physicochemical variables and ionic species concentrations of the SRW. Value pH EC (μS/cm) Dissolved oxygen (%) Dissolved oxygen (mg L−1) DOC (mg L−1) Cl− (mg L−1) −1 NO− ) 2 (mg L −1 NO− ) 3 (mg L PO3− (mg L−1) 4 SO2− (mg L−1) 4 Na+ (mg L−1) −1 NH+ ) 4 (mg L K+ (mg L−1) 2+ −1 Mg (mg L ) Ca2+ (mg L−1) LOD: Limit of detection.

8.23 804.50 102.90 9.18 8.79 ± 0.76 63.34 ± 0.87 bLOD 0.43 ± 0.28 11.05 ± 0.94 99.90 ± 5.52 75.42 ± 1.27 25.39 ± 0.89 9.18 ± 3.48 23.63 ± 0.77 27.41 ± 2.36

Cmax  KL  Cw 1 þ KL  Cw

ð2Þ

where Cs (μg kg−1) and Cw (μg L− 1) are the sorbed and solution concentrations at equilibrium respectively, KL (L μg−1) is the Langmuir constant, and Cmax (μg kg−1) is the maximum sorbed concentration. 2.8. Sorption coefficients of ionisable compounds The widely used distribution coefficient Kd (L kg−1) describes the concentration ratio of the target compounds between the sediment and liquid phases at equilibrium, and is a key variable in the assessment of the mobility and fate of environmental chemicals (Franco and Trapp, 2008). Karickhoff et al. (1979) reported that their Kd values were strongly correlated to the ƒOC (%) in sediment since sorption onto the inorganic fraction was negligible. This correlation is expressed by the normalized organic carbon partition coefficient KOC (L kg−1): KOC ¼

Kd f OC

ð3Þ

The KOC is generally well correlated to the KOW. However, this holds true only for those compounds that are hydrophobic and, therefore, more likely to partition into organic matter or living organisms (bioaccumulation) (Franco and Trapp, 2008; Karickhoff et al., 1979). Since the KOC term for non-hydrophobic and ionisable compounds is pH-dependent (Sabljic et al., 1995), the contribution of the ionised

276

V. Martínez-Hernández et al. / Science of the Total Environment 472 (2014) 273–281

form of the chemicals must be considered. NOM is negatively charged at environmental pH and, therefore, acts as a cation exchanger (Schwarzenbach et al., 2003). Electrostatic interactions of ionised compounds with NOM must therefore be included in the KOC term. The KOC corrected for non-hydrophobic compounds can be calculated according to Eq. (4) (Sabljic et al., 1995): logKOC ¼ 0:52  logOW þ 1:02:

ð4Þ

For ionisable compounds the pH-dependent octanol–water distribution coefficient DOW is used instead of KOW. This takes into account the dissociation constant (pKa), the KOW and solution pH. The DOW was calculated for each target PPCP according to Eqs. (5) and (6), depending on the acidic or alkaline nature of each (Lin et al., 2010; Schaffer et al., 2012b):  logDacid ¼ logKOW þ log

1 1 þ 10pH−pka



  1 : logDbase ¼ logE þ log pka−pH 1 þ 10

ð5Þ

ð6Þ

However, the sorption behaviour of non-hydrophobic, ionisable compounds cannot be fully described by just partitioning and their sorption to organic matter. Other processes involving mineral surfaces have to be considered when determining the Kd. These processes may include ion-exchange, cation bridging, surface complexation and hydrogen bonding (Schaffer et al., 2012a; Schwarzenbach et al., 2003; Tolls, 2001). The experimental Kd can be described as the sum of the contribution of sorption and partitioning into organic matter (KNOM ) plus the d contribution of PPCP sorption to inorganic surfaces (KIS d ) (Schaffer et al., 2012a; Schwarzenbach et al., 2003): NOM

Kd ¼ Kd NOM

Kd

IS

þ Kd



n c ¼ ð1−α Þ  KOC  f OC þ α  KOC  f OC

ð7Þ ð8Þ

where KnOC is the KOC for neutral non-hydrophobic compounds calculated from Eq. (4), KcOC is the KOC for ionised compounds calculated from Eq. (4) but using Dacid or Dbase from Eqs. (7) and (8) instead of KOW, and α is the degree of protonation/deprotonation for bases and acids calculated according to Eqs. (9) and (10) respectively: a¼

1 1 þ 10pH−pka

ð9Þ

a¼

1 1 þ 10pka−pH

ð10Þ

KIS d is calculated by the difference between the experimentally determined Kd and the KNOM . d The pH of the environment may change the surface charge of the sediment. PZC represents the pH at which the amount of cationic sites and anionic sites of a certain adsorbent (in this case the sediment) are equal, providing an overall neutral surface charge. If the solution pH is below the sediment PZC, a positive charge will predominate on the sediment surface; if the solution pH is above the sediment PZC, the sediment surface will be negatively charged. At the experimental pH (7.94 ± 0.17), negatively charged surfaces would predominate in the sediment (PZC = 6.8). Sorption–desorption hysteresis were quantified for each compound at Cw = 100 μg L−1 by the Hysteresis Index (HI) (Huang et al., 1998): des

HI ¼

sor

C s −C s C sor s

ð11Þ

−1 where Cdes and Csor ) s s are the solid phase solute concentration (μg kg for sorption (sor) and desorption (des) experiments at a fixed solution concentration, respectively.

3. Results and discussion 3.1. Sorption of PPCPs onto the sediment The Freundlich and Langmuir models adequately reflected the sorption behaviour of the investigated compounds. Table 3 shows the model variables and the differences between the correlation values (R2) calculated for both models to be minimal. Fig. 1 shows the linearization of the data for the Freundlich isotherm for each PPCP. A linear isotherm results when the Freundlich n value approximates unity. Non-linear sorption was observed for all PPCPs, with n values ranging between 0.37 and 0.87. Results reflecting the non-linear sorption of PPCPs have been reported by other authors (Chefetz et al., 2008; Lin et al., 2010; Ramil et al., 2009). KF values reflect the sorption affinity of the PPCPs for the sediment; however KF units are n-dependent and not comparable if n values are different. Therefore, Kd values were used to compare sorption affinity among the studied PPCPs and those reported in the literature. These Kd values cannot be obtained by fitting the sorption data to the linear equation when non-linearity is observed (Wu et al., 2013; Xu et al., 2009). The present Kd values (Table 3) were therefore calculated as the mean of all the data collected for all the tested concentrations, as proposed by Xu et al. (2009). 3.2. Neutral compounds Negligible sorption onto the sediment was found for ACP and CBZ. The Kd values for ACP and CBZ were 0.5 and 0.4 L kg− 1 respectively. Both compounds were present in their neutral form under the experimental pH. Sorption onto charged surfaces (NOM and/or mineral surfaces) was therefore not expected (Lorphensri et al., 2007; Scheytt et al., 2005). ACP is a weak acid with a pKa of 9.38 and a low logKOW. The logKOW value for CBZ was higher than that for ACP, therefore CBZ would have been more likely partitioned to NOM. However, neither CBZ nor ACP were thus partitioned, nor were they sorbed onto charged surfaces. The calculated KNOM was in both cases greatly overestimated, d and much higher than the experimental Kd (Tables 5 and 6). The negligible sorption affinity for ACP and CBZ would appear to be the result of their low hydrophobicity and/or their uncharged form at pHs typical of natural or reclaimed water. Similar negligible sorption affinity has also been reported by Williams et al. (2009) and Lin et al. (2010). 3.3. Positively charged compounds CAF and ATN were positively ionised at the experimental pH and showed the highest degree of sorption among the investigated PPCPs. Table 3 shows their Kd values. The sediment surface was predominantly negatively charged and therefore would have attracted cations. Of the different sorption mechanisms possible, electrostatic interactions between positively charged compounds and negative charged sediment surfaces (NOM and/or clay mineral surfaces) are probably the most important (Ramil et al., 2009). The estimated KIS d for both compounds IS (KIS d = 0.98 Kd for CAF and Kd = 0.85 Kd for ATN) was significantly higher than the calculated KNOM (Table 4), confirming that interactions d with inorganic surfaces control the sorption of cationic PPCP species onto the sediment. This confirms that cation exchange and electrochemical interactions are important in their sorption. The significance of cation exchange and electrochemical interactions in ATN and CAF sorption, has been also suggested by other authors (Lin et al., 2010; Schaffer et al., 2012b). So far, all studies about sorption of PPCPs measured Kd and only Schaffer et al. (2012a) calculated the KIS d for two

V. Martínez-Hernández et al. / Science of the Total Environment 472 (2014) 273–281

277

Table 3 Sorption model variables and model fitting adjustments. Linear

Freundlich

Langmuir

Compound

Kd

STD

KF

KL

Cmax

R2

n

R2

HI

CBZ ACP ATN CAF NPX SXZ

0.40 0.50 7.93 17.86 1.86 4.25

0.55 0.16 4.01 6.10 1.27 1.61

0.97 0.68 11.00 20.00 3.38 6.20

0.34 0.02 0.08 0.09 0.07 0.03

5.62 32.93 189.01 221.17 68.68 218.78

0.88 0.98 0.99 0.98 0.97 1.00

0.37 0.87 0.76 0.87 0.70 0.79

0.97 0.97 0.99 0.99 0.99 0.99

– – −0.66 −0.30 4.46 13.81

Kd expressed in L kg−1; KF expressed in μg1 − n Ln kg−1; KL expressed in L μg−1; Cmax expressed in μg kg−1. STD: Standard deviation. HI: Hysteresis Index.

β-blockers (ATN and metoprolol). In their study, ATN sorption by cation exchange was higher than 99% of the total sorption. A similar value of Kd for ATN (8.1 L kg−1) was found by Yamamoto et al. (2009). Smaller Kd values (0.75–3.61 L kg−1) were found by Schaffer et al. (2012a) in river sediments. However, these values were obtained by column experiments which involved smaller sediment/ water contact times than batch experiments. In a soil column, water passes through at a finite rate and both reaction time and degree of mixing between water and soil can be much less than those occurring in a laboratory batch test (Krupka et al., 1999). Lin et al. (2010) reported a CAF Kd value one order of magnitude higher than the present finding. The reason for this discrepancy may lie in the fact that Lin et al. (2010) measured their Kd values over a smaller range of concentrations (0.1–6.6 μg L− 1) compared to the present study (1–100 μg L− 1). A CAF Kd value (18.5 L kg− 1) consistent with the present findings has

been reported for similar sandy loam soils (Karnjanapiboonwong et al., 2010). PXT was not detected in any sample, indicating there was no degradation of CAF during the assays. 3.4. Negatively charged compounds SXZ and NPX, which were negatively charged under the present experimental conditions, showed Kd values lower than those obtained for the positively charged PPCPs (Table 3). The estimated KIS d for SXZ was almost three times the calculated KNOM (Table 4), indicating that sorption onto mineral surfaces is d the most important process acting in SXZ retention by the studied sediment. Since the sediment surface was predominantly negatively charged, electrostatic repulsion effects were expected. Ligand exchange

Fig. 1. Sorption and desorption of linearized Freundlich isotherms for the selected PPCPs.

278

V. Martínez-Hernández et al. / Science of the Total Environment 472 (2014) 273–281

Table 4 Sorption coefficients and physicochemical variables for selected compounds. Compound

Character (charge)

pKaa

logKowa

logDow

Knoc

Kcoc

Kd

KNOM d

KIS d

CBZ ACP ATN CAF NPX SXZ

Base (N) Acid (N) Base (+) Base (+) Acid (−) Acid (−)

−0.49b 9.38 9.60 10.40 4.15 1.6; 5.7

2.45 0.46 0.16 −0.07 3.18 0.89

2.45 0.44 −1.51 −2.53 −0.61 −1.35

196.79 18.16 12.68 9.63 471.63 30.39

63.74 5.77 0.56 0.16 1.63 0.67

0.40 0.50 7.93 17.86 1.86 4.25

283.38 25.53 1.18 0.28 2.46 1.21

– – 6.76 17.58 −0.60 3.03

N: Neutral (+): Positively charged (−): Negatively charged. −1 Knoc, Kcoc, KNOM and KIS . d d are expressed in L kg a Hazardous Substance Database (HSDB) from the U.S. National Library of Medicine. TOXNET, 2011. b From Schaffer et al., 2012b.

surface complexation with Altot and/or Fetot oxides on the sediment (Table 1) might therefore occur besides partitioning to NOM. Surface complexation sorption processes for antibiotics and other organic compounds have been reported by other studies (Figueroa and Mackay, 2005; Gu and Karthikeyan, 2005; Pateiro-Moure et al., 2010). As shown in Table 4, the sorption of NPX onto inorganic surfaces was negligible, and partitioning to NOM was the predominant sorption mechanism. This behaviour was predictable based on the high logDOW of NPX (Table 4). Kd values available in the literature for NPX are quite controversial. As suggested by Chefetz et al. (2008) and Durán-Álvarez et al. (2012) sorption of NPX is influenced by the nature of NOM. Indeed, NOM decomposition increases its hydrophobicity due to relatively greater amounts of aromatic and alkyl moieties (Chefetz et al., 2000). A pronounced sorption of NPX onto soil can be due to л–л interactions with aromatic moieties in soil humic substances (Lin and Gan, 2011). As a consequence, NPX Kd values reported in other studies present a high variability (0.20–26.57 L kg−1) (Durán-Álvarez et al., 2012; Williams et al., 2009; Xu et al., 2009). Sorption of PPCPs onto sediment is highly influenced by the type and degree of ionization. Cationic species are more sorbed onto the studied sediment than anionic and neutral compounds. Sorption onto inorganic surfaces was the predominant sorption mechanism for all ionised PPCPs with the exception of NPX. NPX is the most hydrophobic compound among the studied chemicals and this explains its lower sorption onto inorganic surfaces.

2+ respect to the desorption experiment involving SRW, NH+ 4 and Mg continued to be adsorbed by cation exchange with K+ and Ca2+, and to a lesser extent with Na+. Also, PO2− continued to adsorb via surface 4 complexation with the Altot and Fetot oxides of the sediment. These results indicate that the sorption sites of the sediment were not saturated 24 h into the experiment. The negative cationic balances obtained in the sorption and desorption experiments with SRW indicate that cationic species other than those measured were desorbing (Table 6). When the desorption experiments were performed using the CaCl2 solution, all cations (except for Ca2+) previously sorbed to the sediment were desorbed as an effect of equilibration between the sediment and the solution, which contained no ions other than Ca2+ and Cl−. While the sediment continued to desorb large amounts of Mg2 +, Na+ and K+, the NH+ 4 adsorbed during the sorption experiment was not completely desorbed. Thus, under these experimental conditions, its desorption is partially irreversible. This highlights that sorption studies need to be coupled with desorption tests for reliable contaminant mobility assessments to be made. NO− 3 showed no significant sorption–desorption tendency. However, it is well known that NO− 3 does not sorb onto sediment (Almasri and Kaluarachchi, 2007). NO− 2 was never detected during any sorption/ desorption experiment, indicating an absence in all the essays of − any aerobic microbial metabolism able to transform NH+ 4 into NO3 .

3.6. Desorption of PPCPs from the sediment 3.5. Sorption/desorption of inorganic ions in the SRW/CaCl2 onto/from the sediment Table 5 shows the concentrations of inorganic ions recorded in the liquid phases in the sorption and desorption experiments. In the sorption experiments with SRW, significant sorption of PO24 −, NH+ 4 and Mg2+ onto the sediment took place. As shown by the cationic balance calculated for the sorption experiments (Table 6), Mg2 +, NH+ 4 and Na+ sorption occurred by cation exchange with the Ca2+ and K+ of the sediment. PO24 − was probably sorbed by surface complexation with the Altot and Fetot oxides of the sediment (see Table 1). With

The desorption of PPCPs was expected to be influenced by the presence of inorganic ions due to competition for sorption sites (Pateiro-Moure et al., 2007; Sithole and Guy, 1987; Tolls, 2001) and the enhancement of acidic organic compound sorption via the reduction of repulsive electrostatic forces (Tülp et al., 2009). According to Schaffer et al. (2012a), cation competition has been observed between Ca2+ and ATN. Although different behaviours were observed for the inorganic ions during desorption with SRW and CaCl2, no competition between the PPCPs and inorganic ions was found (Fig. 1). The PPCPs showed a similar desorption tendency in the two experiments.

Table 5 Inorganic ion concentrations in the liquid phase. C0 and C24h represent the initial and equilibrium average concentrations in the solution respectively. Standard deviations are shown in brackets. Concentrations are given in mg L−1.

Sorption Desorption

C0 C24h C0 CaCl2 C24h CaCl2 C0 SRW C24h SRW

NO− 3

PO2− 4

SO2− 4

NH+ 4

K+

Na+

Ca2+

Mg2+

0.37 (0.28) 0.86 (0.31) – 0.07 (0.13) 0.65 (0.20) 0.61 (0.10)

11.15 (1.04) 3.30 (0.97) – – 10.70 (0.32) 3.41 (0.50)

101.09 (5.38) 100.96 (9.74) – 15.12 (10.52) 96.23 (3.64) 94.67 (8.77)

25.15 (0.96) 13.04 (1.99) – 7.77 (0.64) 25.42 (1.18) 11.83 (2.03)

9.03 (0.53) 12.33 (0.93) – 10.87 (0.46) 10.08 (0.21) 12.19 (1.07)

75.43 (1.29) 68.07 (4.71) – 15.56 (2.48) 75.48 (1.21) 75.68 (7.66)

27.85 (2.39) 30.26 (5.04) 390.00 (0) 335.16 (7.91) 26.31 (1.64) 30.46 (3.07)

23.71 (0.79) 8.29 (1.18) – 22.09 (1.10) 23.25 (0.62) 11.66 (0.75)

V. Martínez-Hernández et al. / Science of the Total Environment 472 (2014) 273–281 Table 6 Cation balance. Concentrations are given in meq L−1.

Sorption SRW Desorption CaCl2 Desorption SRW

Na+

NH+ 4

K+

Mg2+

Ca2+

Balance

−0.32 0.68 0.01

−0.67 0.43 −0.75

0.05 0.28 0.05

−1.27 1.82 −0.95

0.12 −2.74 0.21

−2.09 0.47 −1.44

Since negligible sorption was observed for the neutral compounds ACP and CBZ, no desorption of these PPCPs was found (or the desorbed concentrations were below the detection limit). No sorption or desorption isotherms are therefore shown. The desorption of the two cationic species, ATN and CAF, was partially reversible (Figs. 1 and 2), that shown by ATN being greater. HI values at Cw = 100 μg L−1 also showed this tendency (Table 3). Fig. 2 shows the sorption, desorption and removal (sorption minus desorption) percentages when the initial PPCP concentrations in the solution were 100 μg L−1. Percentage values given in the white rectangles of Fig. 2 (see desorption column) were calculated from the total concentration sorbed during the sorption test. According to Fig. 2 about 69.4% of the sorbed ATN was desorbed. Thus, an ATN removal of 22.3% was observed. Contrary to the findings of this study, Ramil et al. (2009) reported hysteresis in ATN sorption. This discrepancy is probably related to the fact that the latter authors considered 6 h to be sufficient time for equilibrium to be reached. In the present work, about 20.6% of the sorbed CAF was desorbed (a final retention of more than 60%). These results are similar to those reported by Karnjanapiboonwong et al. (2010) in experiments involving comparable sandy loam soils. The sorption of the anionic compounds SXZ and NPX was hysteretic (more pronounced for SXZ), with the desorption isotherms laying above the sorption isotherms (Fig. 1). Both HI values (Table 3) were higher than zero demonstrating hysteresis in sorption–desorption behaviour. As shown in Fig. 2, only 4.9% of the sorbed SXZ desorbed, indicating a high affinity between the sediment and the compound.

279

Yang et al. (2011) described the reversible sorption of SXZ onto activated sludge. The sorption irreversibility of SXZ in the present study is due to its higher interaction with the inorganic surfaces (KIS d value in Table 4). Bonds between non-hydrophobic compounds and NOM are weaker compared to ligand exchange bonds. NPX desorption was higher than SXZ, with a 31.2% of sorbed compound being desorbed (Fig. 2). As described before, NPX aromatic skeletons can facilitate л–л interactions with aromatic moieties of NOM (Chefetz et al., 2008). л–л interactions are non-covalent bonds and therefore weaker than ligand exchange bonds. This would explain the more pronounced hysteresis of SXZ than that of NPX. A slight dependence of hysteresis on concentration was observed for NPX, suggesting an increase in sorption irreversibility at higher concentrations. These results indicate that PPCP sorption studies have to be coupled with desorption tests to obtain a realistic estimation of retention capacity of sediments. Indeed, ATN which was more sorbed than SXZ was also desorbed more easily resulting in a lower percentage of removal.

4. Conclusions The degree and type of ionisation affect the PPCPs sorption capacity. The cationic PPCPs examined showed a stronger tendency to be sorbed compared to anionic and neutral species. With the exception of NPX, results indicate that sediment inorganic surfaces mainly control sorption of PPCPs. More than 70% of the total sorption is due to interaction with mineral surfaces. This holds especially true for cationic species (ATN and CAF) which sorption is enhanced by the negative surface charge of the sediment. The present results also indicate that the strength of the bond between PPCPs and the sediment surface influences the degree of desorption that may occur. Although different behaviours were observed for the inorganic ions during desorption with SRW and CaCl2, no competition between the PPCPs and inorganic ions was observed. The desorption of the two cationic species, ATN

Fig. 2. Percentage sorption, desorption and retention of the selected compounds at the highest concentration tested (100 μg L−1). The desorption percentages were calculated with respect to sorbed concentrations. Since these percentages do not correspond to the y-axis scale of the graphic, explicit values were given in the white rectangles.

280

V. Martínez-Hernández et al. / Science of the Total Environment 472 (2014) 273–281

and CAF, was partially reversible. The sorption of the anionic compounds SXZ and NPX was hysteretic (more pronounced for SXZ) based on the desorption results. According to sorption/desorption experimental results, the potential of the examined compounds to infiltrate the unsaturated zone without being fully removed and/or becoming available for biota, decreases as follow CBZ N ACP N NPX N ATN N SXZ N CAF. Studying the sorption/desorption behaviour of PPCPs with respect to sediments is important if models that can predict their environmental fate are to be constructed. Acknowledgements This research was partly funded by a Madrid Community Research Grant (CPI/0613/2008), the grant CGL2009-13168-C03-01-02-03 and the Subprogram CONSOLIDER-TRAGUA (CSD2006-00044) from the Spanish Ministry of Science and Innovation. Special thanks go to Francisco Martínez Serrano and Carolina Guillén Fuentes for performing a part of the chemical analyses. References Almasri MN, Kaluarachchi JJ. Modeling nitrate contamination of groundwater in agricultural watersheds. J Hydrol 2007;343:211–29. BOE. Spanish Water Reuse Royal Decree 1620/2007. Ministry of the Presidence; 2007. Chefetz B, Chen Y, Clapp CE, Hatcher PG. Characterization of organic matter in soils by thermochemolysis using tetramethylammonium hydroxide (TMAH). Soil Sci Soc Am J 2000;64:583–9. Chefetz B, Mualem T, Ben-Ari J. Sorption and mobility of pharmaceutical compounds in soil irrigated with reclaimed wastewater. Chemosphere 2008;73:1335–43. Drillia P, Stamatelatou K, Lyberatos G. Fate and mobility of pharmaceuticals in solid matrices. Chemosphere 2005;60:1034–44. Durán-Álvarez JC, Prado-Pano B, Jiménez-Cisneros B. Sorption and desorption of carbamazepine, naproxen and triclosan in a soil irrigated with raw wastewater: estimation of the sorption parameters by considering the initial mass of the compounds in the soil. Chemosphere 2012;88:84–90. Eaton AD, Clesceri LS, Rice EW, Greenberg AE. Standard methods for the examination of water and wastewater. Washington DC (USA): American Public Health Association/American Water Works Association/Water Environment Federation; 2005. Estévez E, Cabrera MC, Molina-Díaz A, Robles-Molina J, Palacios-Díaz MdP. Screening of emerging contaminants and priority substances (2008/105/EC) in reclaimed water for irrigation and groundwater in a volcanic aquifer (Gran Canaria, Canary Islands, Spain). Sci Total Environ 2012;433:538–46. Fenet Hln, Mathieu O, Mahjoub O, Li Z, Hillaire-Buys D, Casellas C, et al. Carbamazepine, carbamazepine epoxide and dihydroxycarbamazepine sorption to soil and occurrence in a wastewater reuse site in Tunisia. Chemosphere 2012;88:49–54. Figueroa RA, Mackay AA. Sorption of oxytetracycline to iron oxides and iron oxide-rich soils. Environ Sci Technol 2005;39:6664–71. Franco A, Trapp S. Estimation of the soil–water partition coefficient normalized to organic carbon for ionizable organic chemicals. Environ Toxicol Chem 2008;27:1995–2004. García-Galán MJ, Garrido T, Fraile J, Ginebreda A, Díaz-Cruz MS, Barceló D. Simultaneous occurrence of nitrates and sulfonamide antibiotics in two ground water bodies of Catalonia (Spain). J Hydrol 2010;383:93–101. Gee WG, Bauder JW. Particle-size analysis. Madison, WI, USA: American Society of Agronomy/Soil Science society of America; 1986. González Alonso S, Catalá M, Maroto RR, Gil JLR, de Miguel ÁG, Valcárcel Y. Pollution by psychoactive pharmaceuticals in the Rivers of Madrid metropolitan area (Spain). Environ Int 2010;36:195–201. Gu C, Karthikeyan KG. Sorption of the antimicrobial ciprofloxacin to aluminum and iron hydrous oxides. Environ Sci Technol 2005;39:9166–73. Huang W, Yu H, Weber WJJ. Hysteresis in the sorption and desorption of hydrophobic organic contaminants by soils and sediments: 1. A comparative analysis of experimental protocols. J Contam Hydrol 1998;31:129–48. Jurado A, Vázquez-Suñé E, Carrera J, López de Alda M, Pujades E, Barceló D. Emerging organic contaminants in groundwater in Spain: a review of sources, recent occurrence and fate in a European context. Sci Total Environ 2012;440:82–94. Karickhoff SW, Brown DS, Scott TA. Sorption of hydrophobic pollutants on natural sediments. Water Res 1979;13:241–8. Karnjanapiboonwong A, Morse A, Maul J, Anderson T. Sorption of estrogens, triclosan, and caffeine in a sandy loam and a silt loam soil. J Soils Sediments 2010;10:1300–7. Katayama A, Bhula R, Burns GR, Carazo E, Felsot A, Hamilton D, et al. Bioavailability of xenobiotics in the soil environment. In: Whitacre DM, editor. Reviews of environmental contamination and toxicology, 203. New York: Springer; 2010. p. 1–86. Kibbey TCG, Paruchuri R, Sabatini DA, Chen LX. Adsorption of beta blockers to environmental surfaces. Environ Sci Technol 2007;41:5349–56. Kiziloglu FM, Turan M, Sahin U, Kuslu Y, Dursun A. Effects of untreated and treated wastewater irrigation on some chemical properties of cauliflower (Brassica olerecea L. var. botrytis) and red cabbage (Brassica olerecea L. var. rubra) grown on calcareous soil in Turkey. Agric Water Manage 2008;95:716–24.

Koeck-Schulmeyer M, Ginebreda A, Postigo C, Lopez-Serna R, Perez S, Brix R, et al. Wastewater reuse in Mediterranean semi-arid areas: the impact of discharges of tertiary treated sewage on the load of polar micro pollutants in the Llobregat river (NE Spain). Chemosphere 2011;82:670–8. Krupka KM, Kaplan DI, Whelan G, Serne RJ, Mattigod SV. Understanding variation in partition coefficient, Kd, values. volume I: the Kd model, methods of measurement, and application of chemical reaction codes. Washington, DC: US Environmental Protection Agency; 1999. Lazarova V, Levine B, Sack J, Cirelli G, Jeffrey P, Muntau H, et al. Role of water reuse for enhancing integrated water management in Europe and Mediterranean countries. Water Sci Technol 2001;43:25–33. Levine AD, Asano T. Peer reviewed: recovering sustainable water from wastewater. Environ Sci Technol 2004;38:201A–8A. Lin K, Gan J. Sorption and degradation of wastewater-associated non-steroidal anti-inflammatory drugs and antibiotics in soils. Chemosphere 2011;83: 240–6. Lin AY-C, Lin C-A, Tung H-H, Chary NS. Potential for biodegradation and sorption of acetaminophen, caffeine, propranolol and acebutolol in lab-scale aqueous environments. J Hazard Mater 2010;183:242–50. Löffler D, Römbke J, Meller M, Ternes TA. Environmental fate of pharmaceuticals in water/sediment systems. Environ Sci Technol 2005;39:5209–18. Lorphensri O, Sabatini DA, Kibbey TCG, Osathaphan K, Saiwan C. Sorption and transport of acetaminophen, 17 alpha-ethynyl estradiol, nalidixic acid with low organic content aquifer sand. Water Res 2007;41:2180–8. Martinez Bueno MJ, Hernando MD, Herrera S, Gomez MJ, Fernandez-Alba AR, Bustamante I, et al. Pilot survey of chemical contaminants from industrial and human activities in river waters of Spain. Int J Environ Anal Chem 2010;90:321–43. McKeague JA, Day JH. Dithionite- and oxalate-extractable Fe and Al as aids in differentiating various classes of soils. Can J Soil Sci 1966;46:13–22. Molinos-Senante M, Hernandez-Sancho F, Sala-Garrido R. Cost–benefit analysis of water-reuse projects for environmental purposes: A case study for Spanish wastewater treatment plants. J Environ Manage 2011;92:3091–7. Nelson DW, Sommers LE. Total carbon, organic carbon and organic matter, vol. 9. Madison, WI, USA: American Society of Agronomy; 1982. Niedbala A, Schaffer M, Licha T, Nödler K, Börnick H, Ruppert H, et al. Influence of competing inorganic cations on the ion exchange equilibrium of the monovalent organic cation metoprolol on natural sediment. Chemosphere 2013;90:1945–51. OECD. Test No. 106: adsorption–desorption using a batch equilibrium method. OECD Publishing; 2000. Pan B, Ning P, Xing BS. Part V. Sorption of pharmaceuticals and personal care products. Environ Sci Pollut Res 2009;16:106–16. Pateiro-Moure M, Perez-Novo C, Arias-Estevez M, Lopez-Periago E, MartinezCarballo E, Simal-Gandara J. Influence of copper on the adsorption and desorption of paraquat, diquat, and difenzoquat in vineyard acid soils. J Agric Food Chem 2007;55:6219–26. Pateiro-Moure M, Bermúdez-Couso A, Fernández-Calviño D, Arias-Estévez M, Rial-Otero R, Simal-Gándara J. Paraquat and diquat sorption on iron oxide coated quartz particles and the effect of phosphates. J Chem Eng Data 2010;55:2668–72. Ramil M, El Aref T, Fink G, Scheurer M, Ternes TA. Fate of beta blockers in aquatic-sediment systems: sorption and biotransformation. Environ Sci Technol 2009;44:962–70. Sabljic A, Güsten H, Verhaar H, Hermens J. QSAR modelling of soil sorption. Improvements and systematics of log KOC vs. log KOW correlations. Chemosphere 1995;31: 4489–514. Schaffer M, Börnick H, Nödler K, Licha T, Worch E. Role of cation exchange processes on the sorption influenced transport of cationic B-blockers in aquifer sediments. Water Res 2012a;46:5472–82. Schaffer M, Boxberger N, Börnick H, Licha T, Worch E. Sorption influenced transport of ionizable pharmaceuticals onto a natural sandy aquifer sediment at different pH. Chemosphere 2012b;87:513–20. Scheytt T, Mersmann P, Lindstadt R, Heberer T. Determination of sorption coefficients of pharmaceutically active substances carbamazepine, diclofenac, and ibuprofen, in sandy sediments. Chemosphere 2005;60:245–53. Schwarzenbach RP, Gschwend PM, Imboden DM. Environmental organic chemistry. Wiley; 2003. Sithole BB, Guy RD. Models for tetracycline in aquatic environments. Water Air Soil Pollut 1987;32:303–14. Stein K, Ramil M, Fink G, Sander M, Ternes TA. Analysis and sorption of psychoactive drugs onto sediment. Environ Sci Technol 2008;42:6415–23. Teijon G, Candela L, Tamoh K, Molina-Díaz A, Fernández-Alba AR. Occurrence of emerging contaminants, priority substances (2008/105/CE) and heavy metals in treated wastewater and groundwater at Depurbaix facility (Barcelona, Spain). Sci Total Environ 2010;408:3584–95. Teijón G, Candela L, Sagristá E, Hidalgo M. Naproxen adsorption–desorption in a sandy aquifer matrix: characterisation of hysteretic behavior at two different temperature values. Soil Sediment Contam Int J 2013;22:641–53. Tolls J. Sorption of veterinary pharmaceuticals in soils: a review. Environ Sci Technol 2001;35:3397–406. Torres T, Maldonado A, Querol R, Zamora I. Miocene alluvial fans of the Madrid basin (Spain), subsurface evolution. Geogaceta 1995;18. Toze S. Reuse of effluent water: benefits and risks. Agric Water Manage 2006;80: 147–59. Trevors JT. Sterilization and inhibition of microbial activity in soil. J Microbiol Methods 1996;26:53–9. Tülp HC, Fenner K, Schwarzenbach RP, Goss K-U. pH-dependent sorption of acidic organic chemicals to soil organic matter. Environ Sci Technol 2009;43:9189–95.

V. Martínez-Hernández et al. / Science of the Total Environment 472 (2014) 273–281 Williams M, Ong PL, Williams DB, Kookana RS. Estimating the sorption of pharmaceuticals based on their pharmacological distribution. Environ Toxicol Chem 2009;28:2572–9. Wu M, Pan B, Zhang D, Xiao D, Li H, Wang C, et al. The sorption of organic contaminants on biochars derived from sediments with high organic carbon content. Chemosphere 2013;90:782–8. Xu J, Wu L, Chang AC. Degradation and adsorption of selected pharmaceuticals and personal care products (PPCPs) in agricultural soils. Chemosphere 2009;77: 1299–305.

281

Yamamoto H, Nakamura Y, Moriguchi S, Nakamura Y, Honda Y, Tamura I, et al. Persistence and partitioning of eight selected pharmaceuticals in the aquatic environment: laboratory photolysis, biodegradation, and sorption experiments. Water Res 2009;43:351–62. Yang S-F, Lin C-F, Yu-Chen Lin A, Andy Hong P-K. Sorption and biodegradation of sulfonamide antibiotics by activated sludge: experimental assessment using batch data obtained under aerobic conditions. Water Res 2011;45:3389–97. Yu Y, Liu Y, Wu LS. Sorption and degradation of pharmaceuticals and personal care products (PPCPs) in soils. Environ Sci Pollut Res 2013;20:4261–7.