Sorption of ionic liquids onto soils: Experimental and chemometric studies

Sorption of ionic liquids onto soils: Experimental and chemometric studies

Chemosphere 88 (2012) 1202–1207 Contents lists available at SciVerse ScienceDirect Chemosphere journal homepage: www.elsevier.com/locate/chemosphere...

441KB Sizes 0 Downloads 53 Views

Chemosphere 88 (2012) 1202–1207

Contents lists available at SciVerse ScienceDirect

Chemosphere journal homepage: www.elsevier.com/locate/chemosphere

Sorption of ionic liquids onto soils: Experimental and chemometric studies Wojciech Mrozik a,⇑, Alicja Kotłowska b, Wojciech Kamysz a, Piotr Stepnowski c a

Department of Inorganic Chemistry, Faculty of Pharmacy, Medical University of Gdansk, al. Hallera 107, 80-416 Gdansk, Poland Department of Food Sciences, Faculty of Pharmacy, Medical University of Gdansk, al. Hallera 107, 80-416 Gdansk, Poland c Department of Environmental Analysis, Faculty of Chemistry, University of Gdansk, ul. Sobieskiego 18, 80-952 Gdansk, Poland b

a r t i c l e

i n f o

Article history: Received 18 January 2012 Received in revised form 20 March 2012 Accepted 24 March 2012 Available online 21 April 2012 Keywords: Ionic liquid Soil Sorption Isotherm Chemometrics Cation exchange capacity

a b s t r a c t Chemometric analyses are a great tool to support typical experimental studies of the interactions of xenobiotics with natural environment. Such interpretations are able to determine statistically significant correlations and finally lead to identification of the major sorption factors. However, to effectively use chemometrics a bigger data set is required. Even though the ionic liquids are intensively studied, their complete fate or prediction of their behavior in the natural environment is still unclear. Therefore, to evaluate and distinguish the patterns of interactions of ILs in soil environment by chemometrics, sorption of nine ionic liquids (imidazolium and pyridinium chlorides) on 11 types of various soils was tested. Experimental studies indicated that compounds with longer alkyl side chains were sorbed far more strongly than weakly lipophilic ones. Moreover, salts with short and/or hydroxylated derivatives were more mobile in soils/sediments and thus, might cause a danger of contamination of surface or ground waters. Cluster analysis revealed that ionic liquids form two major clusters according to interaction with soil surface – one grouping compounds with short and hydroxylated alkyl side chains and the second with the rest of compounds. Pairwise scatterplots for correlations between soil variables and sorption coefficients indicated that the main soil parameter responsible for the sorption was cation exchange capacity. Correlation of sorption coefficients, Kd, with pH indicated the existence of lower sorption potency in lower pH values. Ó 2012 Elsevier Ltd. All rights reserved.

1. Introduction In recent years ionic liquids (ILs) have become a viable alternative to conventional organic solvents. ILs are organic salts which are liquid below 100 °C and may even be liquid at room temperature. A desirable feature of the compounds is their ability to be designed or tuned to a specific chemical process or property such as viscosity, density, miscibility, solubility, etc. by altering cation and/or anion. Moreover, the majority of the ILs is thermally stable, has negligible volatility and is non-flammable (Earle and Seddon, 2000). ILs have found potential application in a variety of industrial processes (Wasserscheid and Welton, 2003; Plechkova and Seddon, 2008); where they are used for instance as media for organic synthesis and biocatalysis, as alternative electrolytes, as phases in separation techniques, for dissolution and recovery of cellulose, and as alternative lubricants. Even though, the commonly quoted negligible volatility of the ILs decreases potential atmospheric contamination, it will lead to increased risk for soil and aqueous environments. In addition, it is possible to design the compounds which are closer to the ideal ‘‘green’’ chemical, while still retaining

⇑ Corresponding author. Tel.: +48 58 3493225; fax: +48 58 3493224. E-mail address: [email protected] (W. Mrozik). 0045-6535/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.chemosphere.2012.03.070

their technological properties. As part of current legislations such as REACH (REACH, 2006) or action plans such as Agenda 21 (Agenda 21, 2004), the fate assessment of newly introduced compounds, such as ILs, in the natural environment and its possible interactions must be characterized. The risk characterization includes determination of mobility, toxicity and contamination potential prior to implementation of a new chemical. The interactions with soil/sediments have a significant role in the fate assessment of the xenobiotics, as they will influence transport, reactivity, and bioavailability of the new compounds. It must be indicated that the ILs commonly used to date are toxic in nature but their toxicities vary considerably across organisms and trophic levels (Latala et al., 2005; Ranke et al., 2007; Stolte et al., 2007; Arning et al., 2008; Łuczak et al., 2010; Phama et al., 2010). In recent years there have been attempts to create different imidazolium, pyridinium or phosphonium compounds substituted with side chains containing e.g. ester, ether, polyether and amide functional groups combined with several anions for better biodegradability (Gathergood and Scammells, 2002; Gathergood et al., 2004; Markiewicz et al., 2009, 2011; Morrissey et al., 2009). Moreover, not only cations must be taken into ecotoxicological consideration but also anions, especially perfluorinated ones (Stolte et al., 2006; Matzke et al., 2007). However, in comparison to the number of cations of ILs, the number of anions used is limited. The research conducted

W. Mrozik et al. / Chemosphere 88 (2012) 1202–1207

on sorption phenomena of ionic liquids (cations) is limited, and is mainly restricted to a small number of chemicals, and soil matrices. Strength of sorption depends on both ionic liquid structure (Stepnowski, 2005) and physicochemical properties of various sorbates, such as cation exchange capacity (CEC), organic matter content (OM) and clay minerals (Gorman-Lewis and Fein, 2004; Stepnowski, 2005; Matzke et al., 2009). For long chained imidazolium and pyridinium salts ‘‘double layer’’ formation phenomena have been observed (Stepnowski et al., 2007). The pore water properties like pH and ionic strength have also been shown to play an important role in the sorption process. Thermodynamic parameters indicate that interactions of ILs with the surface are a spontaneous exothermic process (Mrozik et al., 2008a). Laboratory migration studies through soil layers are in agreement with batch tests showing that longer alkyl chained ILs interact more strongly with the soil surface than the short ones (Mrozik et al., 2009; Studzinska et al., 2009). The attempts to use HPLC to model environmental interactions of ILs and soils were also conducted (Mrozik et al., 2008b) and it was found that such modeling may support the choice of the appropriate test parameters for experimental studies. Despite the performed research, there are still some gaps in understanding of the contribution of the factors responsible for sorption of ILs. The sorption coefficient, Kd, is a summation of the whole sorption process. Therefore, the prediction of the interaction of ILs in the soil matrix is limited, as the number of variables which may contribute to the interaction is large. If the individual contributions to the sorption process may be quantified, the prediction will be improved. The chemometric analysis is a great tool to support experimental studies (Krogh et al., 2008). Such an analysis will be able to determine statistically significant correlations and finally lead to identification of the major sorption factors. However, to effectively use chemometrics a bigger data set is required. Therefore, to evaluate and distinguish the patterns of interactions of ILs in soil environment by chemometrics, the paper will detail the systematic sorption of nine ionic liquids chloride salts (pyridinium and imidazolium cations, also hydroxylated derivatives) on 11 soils with various physiochemical properties. 2. Material and methods 2.1. Chemicals The ionic liquids used in these studies were chlorides of – 1-ethyl (EMIM), 1-butyl – (BMIM), 1-hexyl – (HMIM), 1-methyl-3-octylimidazolium (OMIM), 1-(2-hydroxylethyl)-3-methylimidazolium (EMIMOH), 1-(3-hydroxylpropylyl)-3-methylimidazolium (PMIMOH), N-butyl-4-methylpyridinium (MBPy), N-butylpyridinium (BPy) and N-butyl-4-(dimethyl)aminopyridinium (AmBPy) and were obtained from Merck KGaA (Darmstadt, Germany; purity P98%). The ILs used did not undergo any pretreatment. Trifluoroacetic acid (TFA) was obtained from Sigma-Aldrich (USA), HPLC gradient grade acetonitrile from Lab-Scan (Ireland) and anhydrite calcium chloride was purchased from P.P.U. ‘‘Standard’’ (Poland). 2.2. Soils The soils were obtained from the various places in Poland and were typical representatives of their type. Detailed information of the origin and physicochemical properties of the soils are presented in Table 1. All samples were air-dried, ground, and sieved. Prior to analysis, soils were characterized by: cation exchange capacity (CEC) (Gillman and Sumpter, 1986), organic matter content (OM) by loss on ignition; pH was determined with a glass elec-

1203

trode in a 1:2.5 soil/water suspension using deionized water and a 1 M KCl solution; The grain size distribution of the soils was determined using a sieve shaker (sand) and by pipette analysis (silt and clay). 2.3. Adsorption of ionic liquids The batch-equilibrium technique (OECD, 2000) was used to determine sorption capacities of the ionic liquids. Various concentrations of ionic liquids (ten concentrations in range of 0.1–200 mM for OMIM (Jungnickel et al., 2008) and 17 concentrations in range of 0.1–500 mM for the rest of compounds) prepared in 0.01 M CaCl2 were added (5 mL) to 1 g of the soil. The mixtures were agitatively, horizontally shaken for 24 h, and then centrifuged for 10 min at 3.075g. The aqueous phase was recovered by filtration (0.45 lm) as completely as possible, and the amount of sorbed ionic liquid was determined by direct analysis of the supernatant by HPLC according to the method described below. Samples were diluted if needed. All experiments were performed in triplicate. The ratio of the amount of ionic liquid sorbed by the soil to the amount of the analyte in the solution at equilibrium state was calculated as sorption coefficient, Kd. 2.4. Ionic liquid analysis To analyze the ionic liquids we used high-performance liquid chromatography (Perkin Elmer Series 200) with a Phenomenex Gemini C6-Phenyl 5 l, 110 Å, 150  4.6 mm LC column (Torrance, USA); mobile phase consisted of acetonitrile and 0.1% TFA in water (5:95–27:73 v/v), pH 3, at an isocratic flow rate of 0.8 mL min1. The elution profiles were monitored at a wavelength k = 218 nm. The method was selective and reliable as proven in previous publications (Stepnowski et al., 2007; Mrozik et al., 2008b, 2009). 2.5. Chemometric studies The statistical interpretation of the results was carried out using Excel 2007 (Microsoft Corp., Seattle, USA) and Statistica 7.1 (Statsoft, Tulsa, USA). Investigation of the distribution of analytes was performed by applying Kolgomorov–Smirnov normality test. The data were distributed normally, thus no transformation was used. All quantitative results were normalized prior to analyses. Visualization of the differences in sorption of ionic liquids was obtained by subjecting the data to hierarchical cluster analysis (HCA) using Euclidean distance and Ward’s method of aggregation. Principal component analysis (PCA) based on correlation matrix was also carried out to group examined analytes and verify the statistical significance of the differences in sorption and desorption coefficients. Prediction of soil influence on the sorption of ionic liquids was performed by analyzing the correlations between soil variables. Statistical significance was considered at P < 0.05. 3. Results and discussion Fig. 1 presents exemplary isotherms for selected examined ILs and soils. Tables 2 and 3 list sorption coefficients and Freundlich isotherm parameters, respectively. Obtained experimental results are in close relation with previous studies (Stepnowski, 2005; Stepnowski et al., 2007; Beaulieu et al., 2008; Studzinska et al., 2008; Mrozik et al., 2009) in accordance to double layer sorption mechanism. However, for some soils the final saturation of the second layer was not observed, especially for (OMIM), it was also hardly observed for short chain compounds. Moreover, the influence of structure of ionic liquids on strength of sorption was ob-

1204

W. Mrozik et al. / Chemosphere 88 (2012) 1202–1207

Table 1 Soil characteristics. Type of soil

ID

Agricultural chernozem

R1

Clayey brown soil

R2

Alluvial agricultural soil

R3

Loessy Brown soil

R4

Acidic Brown soil

L1

Forest Podsolic soil

L2

Sandy Brown soil

M1

Loess soil

M2

Sandy–clayey silt

CA1

Glacial till

CA2

Beach sand

CA3

Sampling area 0

00

50°11 01 N 17°010 4500 E 54°130 0000 N 19°320 4100 E 54°150 1500 N 18°550 5000 E 50°590 3700 N 16°120 0000 E 51°020 1100 N 15°190 1500 E 54°250 0900 N 18°310 1500 E 51°030 3400 N 15°100 1800 E 54°260 0100 N 18°280 4800 E 54°290 0500 N 18°340 1300 E 54°290 0500 N 18°340 1300 E 54°290 0500 N 18°340 1300 E

Fig. 1. Exemplary isotherm of (OMIM) and (EMIMOH) on soil CA2.

served – the longer the side chain the stronger the binding and the general order of intensity was as follows: (OMIM) > (HMIM) > (BMIM) > (EMIM), (PMIMOH) > (EMIMOH). The highest sorption coefficients, Kd, were observed for (OMIM) on R3 soil (12.1 mL g1), the weakest for (EMIMOH) on CA3 soil (0.4 mL g1). Alkylpyridinium salts also exhibited the correlation of structure to sorption strength. The highest Kd for (AmBPy) was for R3 soil (23.7 mL g1), what was also the highest value for all ionic liquids tested (both imidazolium and pyridinium salts). It is due to the two

pHKCl

OM (%)

CEC (meq g1)

Clay content (%)

7.2

20.5

207

57.1

5.8

6.0

99

69.3

6.6

5.5

298

60.5

6.5

6.5

78

47.4

5.2

7.5

70

71.2

4.3

3.9

48

28.2

5.4

7.3

100

73.8

5.3

4.6

130

57.9

5.3

21.5

270

94.0 62.3

7.7

1.54

64

7.6

0.14

30

0.17

positive charges in the structure of this IL. The weakest sorption was noticed for (BPy) on R4 and CA2 soils (0.7 mL g1). An interesting effect was observed for CA3 soil, which was a low sorption potency sorbent (very low in CEC and almost lack of OM or clay minerals). Probably, it was due to a small number of interaction sites or the fact that they were hard to reach. Therefore, salts like (AmBPy), which are more structurally complex, are less sterically favorable to such sites what results in lower Kd values in comparison to (OMIM). However, a second sorption layer was formed – what indicates of interactions among ILs particles. Higher desorption rates were typical for salts with lower sorption potency (shorter alkyl side chains). Moieties of (EMIM) or (BPy) were easily removed from soil surface and may have potential to penetrate to deeper soil profiles/aquifers or be flushed with run-off. However, on weakly sorbing soils (CA2, CA3 and L2) even long alkyl-chain salts ((HMIM), (OMIM) and (AmBPy)) were much more prone to desorption. Low CEC and lack of organic matter in those soils result only in dispersive interactions among sorbed ILs, which are easy to break. Recovery of hydroxylated salts was very high, even up to 98% from CA-3. The differences in the behavior of ionic liquids in respect to soil type were visualized using HCA and PCA. The analyzed compounds were organized in such a way that objects from a given cluster possessed greatest similarity. The dendrogram depicting clustering of ionic liquids using both methods is shown in Fig. 2. In our study a distinct separation of (EMIMOH), (EMIM) and (PMIMOH) was observed in cluster B with respect to other examined ionic liquids. Separation of the rest of the compounds was less distinct and the ILs formed a second cluster with three subclusters (cluster A). Such behavior was probably based on the structure of ionic liquids. Presence of short or/and hydroxylated side chain drastically lowers the potency of strong and permanent binding to soils. Pyridinium salts form one of the subclusters, what on one hand is obvious regarding the chemical structure but on the other hand experimental Kd values show higher similarity between (BMIM) and (MBPy). Freundlich isotherm was used to describe experimental data, because this model fits better than Langmuir, to heterogeneous surfaces like soils (Schwarzenbach et al., 2003). Freundlich constant, KF, indicated that structure of ionic liquids (especially length of alkyl side chain) was a major factor determining the strength of binding to soil surface. For all soils, except for CA, the slope (1/n)

1205

W. Mrozik et al. / Chemosphere 88 (2012) 1202–1207 Table 2 Sorption coefficients. Kd (mL g1) for tested compounds. Ionic liquid chlorides

R1

R2

R3

R4

L1

L2

M1

M2

CA1

CA2

CA3

(EMIM) (EMIMOH) (PMIMOH) (BMIM) (HMIM) (OMIM) (BPy) (MBPy) (AmBPy)

2.7 ± 0.12 1.7 ± 0.15 0.7 ± 0.06 6.2 ± 0.31 6.9 ± 0.45 7.1 ± 0.49 8.0 ± 0.17 10.2 ± 0.80 10.2 ± 0.92

2.6 ± 0.13 2.1 ± 0.19 2.3 ± 0.19 3.7 ± 0.21 4.2 ± 0.25 10.6 ± 0.75 4.1 ± 0.13 5.5 ± 0.37 8.6 ± 0.77

2.3 ± 0.10 2.5 ± 0.23 2.9 ± 0.23 4.6 ± 0.23 7.5 ± 0.49 12.1 ± 0.83 4.4 ± 0.09 6.4 ± 0.50 23.7 ± 2.61

2.0 ± 0.16 2.0 ± 0.17 2.0 ± 0.15 2.7 ± 0.15 3.6 ± 0.18 5.1 ± 0.36 0.7 ± 0.02 1.5 ± 0.08 3.7 ± 0.37

1.9 ± 0.19 0.7 ± 0.01 1.1 ± 0.03 5.3 ± 0.19 5.8 ± 0.15 6.8 ± 0.13 1.7 ± 0.07 5.1 ± 0.36 1.9 ± 0.15

0.7 ± 0.03 0.9 ± 0.08 0.7 ± 0.06 1.0 ± 0.05 2.2 ± 0.14 1.7 ± 0.12 0.9 ± 0.02 1.2 ± 0.09 5.1 ± 0.46

1.4 ± 0.06 1.8 ± 0.16 2.5 ± 0.20 4.4 ± 0.22 7.8 ± 0.51 5.9 ± 0.41 1.4 ± 0.03 10.1 ± 0.79 3.7 ± 0.41

1.9 ± 0.15 1.4 ± 0.12 0.9 ± 0.07 2.4 ± 0.14 5.3 ± 0.27 6.4 ± 0.45 1.8 ± 0.06 2.4 ± 0.13 8.0 ± 0.80

1.3 ± 0.05 1.0 ± 0.01 2.1 ± 0.19 6.8 ± 0.51 9.0 ± 0.38 8.6 ± 0.77 2.6 ± 0.03 3.1 ± 0.22 4.9 ± 0.32

2.0 ± 0.09 2.0 ± 0.18 2.0 ± 0.16 2.7 ± 0.14 3.6 ± 0.23 5.1 ± 0.35 0.7 ± 0.01 1.5 ± 0.12 3.7 ± 0.33

0.5 ± 0.04 0.4 ± 0.03 1.1 ± 0.08 0.9 ± 0.05 2.0 ± 0.10 3.7 ± 0.26 1.2 ± 0.04 1.3 ± 0.07 1.6 ± 0.16

Table 3 Freundlich isotherm parameters for examined soils and ionic liquids. Ionic liquid

Soil R1

[EMIM][Cl] [EMIMOH][Cl] [PMIMOH][Cl] [BMIM][Cl] [HMIM][Cl] [OMIM][Cl] [BPy][Cl] [MBPy][Cl] [AmBPy][Cl]

R2

R4

L1

L2

KF

R2

1/n

KF

R2

1/n

KF

R2

1/n

KF

R2

1/n

KF

R2

1/n

KF

R2

0.81 0.98 0.69 0.83 0.79 0.77 0.85 0.81 0.76

53.1 5.9 20.0 67.7 50.5 51.9 37.3 71.9 109.7

0.974 0.990 0.980 0.986 0.995 0.995 0.999 0.986 0.994

0.75 0.88 0.59 0.86 0.84 0.88 0.91 1.02 1.28

16.7 10.6 12.7 19.5 20.2 20.8 11.4 5.4 21.3

0.987 0.988 0.905 0.990 0.978 0.996 0.998 0.988 0.985

0.63 0.70 0.78 0.62 0.63 0.74 0.73 0.73 0.98

188.8 43.8 41.1 238.7 348.3 154.8 80.1 100.0 35.6

0.978 0.933 0.97 0.990 0.957 0.970 0.995 0.997 0.996

0.85 0.79 0.68 0.87 0.89 0.80 0.55 0.83 0.52

5.9 9.9 34.4 7.3 10.7 30.1 94.6 20.6 94.6

0.974 0.999 0.946 0.973 0.988 0.991 0.972 0.963 0.972

0.87 0.89 0.83 0.90 0.70 0.81 0.78 0.89 0.67

4.62 9.80 4.2 19.5 5 28.5 23.9 20.7 75.6

0.971 0.992 0.985 0.996 0.995 0.987 0.988 0.993 0.978

0.71 0.81 0.72 0.76 0.93 0.98 0.65 0.66 0.89

33.4 6.6 17.5 11.0 4.1 29.9 37.3 49.7 12.9

0.976 0.988 0.995 0.984 0.969 0.961 0.973 0.992 0.937

1/n

KF

R2

1/n

KF

R2

1/n

KF

R2

1/n

KF

R2

1/n

KF

R2

0.71 0.58 0.78 0.96 0.86 0.69 0.71 0.89 0.96

3.4 10.7 14.8 27.1 45.7 65.6 76.5 47.9 19.6

0.976 0.987 0.920 0.995 0.963 0.989 0.991 0.996 0.995

0.56 0.55 0.62 0.58 0.70 0.68 0.57 0.57 0.94

165.5 03.3 43.2 195.9 63.4 89.9 175.4 174.5 17.5

0.980 0.957 0.974 0.983 0.979 0.989 0.946 0.973 0.991

0.66 0.64 0.68 0.68 0.67 0.67 0.60 0.61 0.63

36.4 37.6 35.6 96.8 175.4 248.9 172.9 157 147.6

0.986 0.987 0.978 0.989 0.981 0.959 0.967 0.947 0.989

0.66 0.69 0.73 0.68 0.67 0.53 0.62 0.64 0.69

25.0 23.7 18.5 96.8 74.1 234 69.6 58.9 51.2

0.978 0.987 0.990 0.989 0.992 0.880 0.949 0.954 0.941

0.67 0.65 0.70 0.71 0.73 0.92 0.57 0.60 0.62

19.9 20.0 16.2 20.3 20.9 26.4 57.1 46.0 14.3

0.960 0.956 0.968 0.973 0.950 0.910 0.936 0.931 0.940

M1

[EMIM][Cl] [EMIMOH][Cl] [PMIMOH][Cl] [BMIM][Cl] [HMIM][Cl] [OMIM][Cl] [BPy][Cl] [MBPy][Cl] [AmBPy][Cl]

R3

1/n

M2

CA1

CA2

CA3

Fig. 3. Hierarchical clustering of examined soils. Fig. 2. Hierarchical clustering of examined ionic liquids.

was in the range of 0.7–1, what exhibits concentration-dependent sorption. For CA soils it was slightly lower (0.6–0.7). Sorption of ionic liquids on soil surface is strongly related to physical and chemical properties of the soil itself. The dendrogram (Fig. 3) of the Kd values for each soils shows two major clusters. Cluster A groups the most strongly sorbing soils (R3, R1 and CA1).

Cluster B consists of two subclusters which group semi-strength sorbing soils and weakly sorbing ones. Thus, the order of sorption strength for soil may be presented as follows: R3 > R1 > CA1 > M2 > M1 > R2 > L1 > R4 > CA2 > L2 > CA3. Pairwise scatterplots for correlations between soil variables and the correlation matrix obtained for all the data in the study are shown in Fig. 4. Obtained results indicate the relation of the strength of sorption of ILs to CEC of soil. The highest correlations could be observed for

1206

W. Mrozik et al. / Chemosphere 88 (2012) 1202–1207

CEC (0.81) – what is convergent with the experimental data – in soils with higher the CEC, and where the Kds were also high (0.63) desorption hardly occurred. To have a better insight into the interactions of single structure of ILs with soil matrix, correlations only for (OMIM) and (EMIMOH) (data not shown) were prepared. Molecules with long substituent have correlation of values (spacja podwojna) closer to these presented on the Fig. 1, for CEC (0.74), OM (0.33) and clay content (0.60). Short substituted salt gave weaker correlations (0.15, 0.28 and 0.04, respectively), what again indicates on weak retention of such structure. 4. Conclusion

Fig. 4. Pairwise scatterplots for correlations between soil variables and the correlation matrix. The pH expressed as concentration of [H+] ions.

Kd and CEC, what may be relevant regarding ionic structure of the ILs. Soil R3 possessed the greatest CEC value and CA3 the lowest. It means the interactions of ILs were present due to large number of negative charges on soil surface, seen as active sites. Numerous sites result in a greater number of bonded particles what allows the formation of the second layer (by dispersive interaction). Cation exchange capacity, which results both from clay minerals and organic matter content, gave higher value (0.83) than these both compartments alone (0.55 and 0.49 to Kd, respectively) and that might indicate that clay minerals are a mix of various types (i.e: illite, kaolinite, etc.) and/or organic matter was probably not so well humificated. However, correlation of OM with CEC was higher than clay minerals with CEC (0.66–0.54, respectively) and this indicated that in the case of soils tested more interaction sites come from organic matter. Apart from ionic interactions, organic matter may interact via other interactions such as: van der Waals, dispersive, hydrogen bonding, charge transfer, hpii–hpii interactions, etc. Thus, sorption processes are both adsorption onto and partition into the OM. However, not only quantity but also the quality of organic matter is important. It is said that fulvic acids are an earlier stage of humification of humic acids (Kumada, 1987). In humic acids (HA) carbon to oxygen ratio is 1:1.3 (Sposito, 2008) whereas in fulvic acid (FA) it is 1:1.5 and therefore FA possess lower sorption potential. In our study soils R1 and CA1 exhibit relatively large OM content (20.5% and 29.8%, respectively) but the highest sorption potential is observed for soil R3 with only 5.49%. That may prove very good humification in R3 soil. General correlation of sorption coefficient with pH (as log H+) is poor what may seem strange as this property influences strongly the number of negative charged sites on soil surface. However, due the fact pH is a log value, we have expressed this property of the soil as H+ concentration. Obtained correlation with Kd was drastically higher and negative in nature (0.34). This meant that with lower values of pH (higher concentration of hydrogen ions), the weaker binding of ILs cations was observed. The increase in pH resulted in a greater number of interaction sites (negative charge on surface) for cations. Soils R3, R1 and CA1 have the highest pH what is reflected in stronger sorption phenomena. The Freundlich constant KF is also well correlated with CEC (0.83) but more weakly (than Kd) with clay content and organic matter (0.40 and 0.30, respectively). The mutual correlation between KF and Kd is 0.51, what may be explained by the fact that Freundlich constant is a more general value, whereas sorption coefficient is the sum of several sorption mechanisms, namely adsorption and absorption to natural organic matter, interactions with mineral surfaces, bonding to several different kinds of surface moieties, and finally, interactions of charged species with solids (Schwarzenbach et al., 2003). Desorption is strongly (negative value) correlated to

Mobility and transformation of the ionic liquids depends on parameters of the environmental matrix and its own properties. So far, there was no data published that claimed the presence ILs already in natural environment. However, growing interest in industrial application will probably result in their uncontrolled transfer. Apart from experimental studies, we could find some analogical behavior in organic chemicals, which are already in use and compare it to ILs. The group possessing the biggest similarity are bipirydinium herbicides (paraquat or diquat). Even though the sorption coefficients for these ionic pesticides are far much higher (Bromilow, 2004), the long-alkyl-chain-length ILs are expected to interact with soils/sediments organic matter quite strongly thus being rather immobile. On the other hand, weakly lipophilic salts are suspected to be far much mobile in soils/sediments and may cause the danger of contamination of surface or ground waters. Weaker sorption may also results in possible higher bioavailability. Chemometric studies revealed that the most important property of soil is its cation exchange capacity. Most environmental surfaces (soils, sediments, clay minerals but also bacteria) are negatively charged favoring interactions with ionic liquids. However, the total CEC is a sum of clay minerals and organic matter, what may also enhance other interactions like dispersive ones. Statistical analysis may be a supporting tool in evaluating experimental sorption data. As the Kd is general value, which is the sum of several sorption mechanisms (adsorption and absorption to OM, interactions with mineral surfaces, etc.) (Schwarzenbach et al., 2003), chemometric analysis will allow to determine a more detailed relationship between the strength of interaction with various environmental matrices of chemicals. Such approach may be adopted for chemicals, similar in structure to ILs, such as surfactants, to assess which physicochemical parameters of surfactants (hydrophobicity, specific hydrophobic surface area of surfactants or their capacity to interact with various surfaces) or any additional factors (presence of salt, changes in pH) are responsible for strength of xenobiotics sorption. The future work should focus on detailed explanations of sorption mechanism both with different natural environmental surfaces as well as with artificial sorbents. Moreover, the rate of ILs biodegradation and transformation in natural environment must also be taken into consideration. Acknowledgements The authors would like to acknowledge the financial support of the Polish Ministry of Research and Higher Education under Grant NN 204 527139 and DS 8200-4-0085-1. References Agenda 21, 2004. Environmentally sound management of toxic chemicals, including prevention of illegal international traffic in toxic and dangerous products. In: UN Department of Economic and Social Affairs (Ed.). DfSD.

W. Mrozik et al. / Chemosphere 88 (2012) 1202–1207 Arning, J., Stolte, S., Böschen, A., Stock, F., Pitner, W.R., Welz-Biermann, U., Jastorff, B., Ranke, J., 2008. Qualitative and quantitative structure activity relationships for the inhibitory effects of cationic head groups, functionalised side chains and anions of ionic liquids on acetylcholinesterase. Green Chem. 10, 47–58. Beaulieu, J.J., Tank, J.L., Kopacz, M., 2008. Sorption of imidazolium-based ionic liquids to aquatic sediments. Chemosphere 70, 1320–1328. Bromilow, R.H., 2004. Paraquat and sustainable agriculture. Pest Manage. Sci. 60, 340–349. Earle, M.J., Seddon, K.R., 2000. Ionic liquids. Green solvents for the future. Pure Appl. Chem. 72, 1391–1398. Gathergood, N., Garcia, M.T., Scammells, P.J., 2004. Biodegradable ionic liquids: Part I. Concept, preliminary targets and evaluation. Green Chem. 6, 166–175. Gathergood, N., Scammells, P.J., 2002. Design and preparation of roomtemperature ionic liquids containing biodegradable side chains. Aust. J. Chem. 55, 557–560. Gillman, G.P., Sumpter, E.A., 1986. Modification of the compulsive exchange method for measuring exchange characteristics of soils. Aust. J. Soil Res. 24, 61–66. Gorman-Lewis, D.J., Fein, J.B., 2004. Experimental study of the adsorption of an ionic liquid onto bacterial and mineral surfaces. Environ. Sci. Technol. 38, 2491–2495. Jungnickel, C., Łuczak, J., Ranke, J., Fernández, J.F., Müller, A., Thöming, J., 2008. Micelle formation of imidazolium ionic liquids in aqueous solution. Colloids Surf. A 316, 278–284. Krogh, K.A., Søeborg, T., Brodin, B., Halling-Sørensen, B., 2008. Sorption and mobility of ivermectin in different soils. J. Environ. Qual. 37, 2202–2211. Kumada, K., 1987. Chemistry of Soil Organic Matter. Japan Scientific Societies Press Elsevier, Tokyo. Latala, A., Stepnowski, P., Nedzi, M., Mrozik, W., 2005. Marine toxicity assessment of imidazolium ionic liquids: acute effects on the Baltic algae Oocystis submarina and Cyclotella meneghiniana. Aquat. Toxicol. 73, 91–98. Łuczak, J., Jungnickel, C., Łacka, I., Stolte, S., Hupka, J., 2010. Anti-microbial and surface activity of 1-alkyl-3-methylimidazolium derivatives. Green Chem. 12, 593–601. Markiewicz, M., Jungnickel, C., Markowska, A., Szczepaniak, U., Paszkiewicz, M., Hupka, J., 2009. 1-Methyl-3-octylimidazolium chloride – sorption and primary biodegradation analysis in activated sewage sludge. Molecules 14, 4396–4405. Markiewicz, M., Stolte, S., Lustig, Z., Łuczak, J., Skup, M., Hupka, J., Jungnickel, C., 2011. Influence of microbial adaption and supplementation of nutrients on the biodegradation of ionic liquids in sewage sludge treatment processes. J. Hazard. Mater. 195, 378–382. Matzke, M., Stolte, S., Thiele, K., Juffernholz, T., Arning, J., Ranke, J., Welz-Biermann, U., Jastorff, B., 2007. The influence of anion species on the toxicity of 1-alkyl-3methylimidazolium ionic liquids observed in an (eco) toxicological test battery. Green Chem. 9, 1198–1207. Matzke, M., Thiele, K., Müller, A., Filser, J., 2009. Sorption and desorption of imidazolium based ionic liquids in different soil types. Chemosphere 74, 568–574.

1207

Morrissey, S., Pegot, B., Coleman, D., Garcia, M.T., Ferguson, D., Quilty, B., Gathergood, N., 2009. Biodegradable, non-bactericidal oxygen-functionalised imidazolium esters: a step towards ‘greener’ ionic liquids. Green Chem. 11, 475–483. Mrozik, W., Jungnickel, C., Pitner, W.R., Kumirska, J., Kaczyn´ski, Z., Stepnowski, P., 2009. Predicting mobility of alkylimidazolium ionic liquids in soils. J. Soils Sediments 9, 237–245. Mrozik, W., Jungnickel, C., Skup, M., Urbaszek, P., Stepnowski, P., 2008a. Determination of the adsorption mechanism of imidazoliumtype ionic liquids onto kaolinite: implications for their fate and transport in the soil environment. Environ. Chem. 5, 299–306. Mrozik, W., Nichthauser, J., Stepnowski, P., 2008b. Prediction of the adsorption coefficients for imidazolium ionic liquids in soils using cyanopropyl stationary. Pol. J. Environ. Stud. 17, 383–388. OECD, 2000. OECD Guideline for testing of chemicals. In: OECD (Ed.). 106. Phama, T.P.T., Choa, C.-W., Yun, Y.-S., 2010. Environmental fate and toxicity of ionic liquids: a review. Water Res. 44, 352–372. Plechkova, N.V., Seddon, K.R., 2008. Applications of ionic liquids in the chemical industry. Chem. Soc. Rev. 37, 123–150. Ranke, J., Stolte, S., Störmann, R., Arning, J., Jastorff, B., 2007. Design of sustainable chemical products – the example of ionic liquids. Chem. Rev. 107, 2183–2208. REACH, 2006. Regulation (EC) No. 1907/2006 of the European parliament and of the council. Official Journal of the European Union, pp. 1–849. Schwarzenbach, R.P., Gschwend, P.M., Imboden, D.M., 2003. Environmental Organic Chemistry, second ed. John Wiley & Sons, Inc., Hoboken, New Jersey. Sposito, G., 2008. The Chemistry of Soils, IIed ed. Oxford University Press, Oxford. Stepnowski, P., 2005. Preliminary assessment of the sorption of some alkyl imidazolium cations as used in ionic liquids to soils and sediments. Aust. J. Chem. 58, 170–173. Stepnowski, P., Mrozik, W., Nichthauser, J., 2007. Adsorption of alkylimidazolium and alkylpyridinium ionic liquids onto natural soils. Environ. Sci. Technol. 41, 511–516. Stolte, S., Arning, J., Bottin-Weber, U., Matzke, M., Stock, F., Thiele, K., Uerdingen, M., Welz-Biermann, U., Jastorff, B., Ranke, J., 2006. Anion effects on the cytotoxicity of ionic liquids. Green Chem. 8, 621–629. Stolte, S., Arning, J., Bottin-Weber, U., Müller, A., Pitner, W.R., Welz-Biermann, U., Jastorff, B., Ranke, J., 2007. Effects of different head groups and functionalised side chains on the cytotoxicity of ionic liquids. Green Chem. 9, 760–767. Studzinska, S., Kowalkowski, T., Buszewski, B., 2009. Study of ionic liquid cations transport in soil. J. Hazard. Mater. 168, 1542–1547. Studzinska, S., Sprynskyy, M., Buszewski, B., 2008. Study of sorption kinetics of some ionic liquids on different soil types. Chemosphere 71, 2121–2128. Wasserscheid, P., Welton, T. (Eds.), 2003. Ionic Liquids in Synthesis. Wiley-VCH.