Toxicity and biodegradability of dicationic ionic liquids

Abstract

Ionic liquids (ILs) formed by multivalent cations are generally of higher thermal and electrochemical stability, which makes them attractive for use in high-temperature applications. Whereas the influence of structural elements on the physicochemical properties of dicationic ILs (DILs) is well established, such systematic investigations on their ecotoxicity and biodegradablility are still lacking. The present study investigates the influence of the dicationic structural elements on these characteristics and addresses the question whether already established structure–activity relationships of common ILs can be applied to DILs. Therefore, a set of 10 DILs with different linkage chain length, terminal alkyl side chain length, linkage chain polarity and head groups were synthesized and studied in several biodegradation and toxicity tests. The results showed that the acute toxicity was in many cases below the levels observed for monocationic ILs. However, none of the DILs could be degraded within the performed biodegradation experiments. Hence, DILs are a potential less toxic alternative to monocationic ILs, but further work on their design is necessary.

---

1 Introduction

Ionic liquids (ILs), salts with low melting points, have been increasingly studied in recent years. They are mostly composed of a bulky organic cation and an inorganic or organic anion. Based on the appropriate combination of these structural elements, ILs can be designed with desired properties and possess unique properties like high thermal stability, a wide liquid range or a broad electrochemical window.

ILs are applicable in various fields, e.g. as solvent,¹ catalyst,² in lithium ion batteries³ or dye-sensitized solar cells.⁴ Some ILs are already used in industrial processes.⁵ In terms of the “Green Chemistry concept”, ILs have the advantage of enhanced operational safety, based on their low vapor pressure and therefore lower evaporation rate. However, their toxicity and the environmental hazards that they could pose are also of the highest importance. Many reports on the influence of the cationic head group, length and functionalization of the side chain and type of anion have been published in the last 10 years. The results are summarized in various review articles.^6–8 In brief, ILs exhibit a varying hazard potential, depending on the structural elements they consist of. Regardless of the investigated test system, lipophilicity appears to have the highest impact on such a compound's toxicity. In biodegradation studies, ILs with long and/or functionalized (e.g. alcohol or carboxylic acid) alkyl chains often show better biodegradation rates.⁹

Recently, multivalent cations have also been used to form ILs.^10,11 Here two or more cationic head groups are linked by alkyl chain(s). Because it is possible to form either symmetrical or unsymmetrical ILs, containing different head groups and substituents, the number of compounds and the design potential is further enlarged. Compared to monocationic ILs, multicationic ones can have a higher melting point, viscosity, surface tension, thermal stability and a wider liquid range.^10,12–14 The higher thermal stability in particular makes them attractive for use as high-temperature lubricants^14–16 or as the stationary phase in gas chromatography.^17,18 They have also been used as electrolytes in secondary batteries¹⁹ and in dye-sensitized solar cells.^20,21 The influence of structural elements like the type of head group and anion, length of side or linkage chain and symmetry on the physicochemical properties of ILs has already been studied.^{10,14,22–24} Such systematic studies of their environmental impact are still lacking. To the best of our knowledge, examinations have just been done for selected representatives. 1,3-Bis[4-methylpyridinium] dibromide, for instance, was found to own a lower microbial activity among all tested species (Gram-positive and Gram-negative bacteria and yeast) compared to the tested monocations.²⁵ However, substitution of the anions to tetrafluoroborate led to compounds with similar growth inhibition as traditional ILs.²⁵ Ford et al. showed that an acetal-linked bis-pyridinium IL was not biodegraded in a CO₂ headspace test (OECD 310).²⁶

In this study, we focused on whether the dicationic structural element changes the toxicity and/or the biodegradability in comparison with traditional IL structures. Moreover, we wanted to address the question whether already established structure–activity relationships of monocationic ILs could be applied to dicationic ILs (DILs). To do so, we synthesized a set of 10 DILs (Table 1) in order to perform a systematic investigation into the influence of:

Table 1 Results from toxicity tests for DILs and monocationic ILs

No.	Structure		IC50 [μM]^a		EC50 [μM]^b
			AChE	IPC-81	S. vacuolatus	D. magna
a Half maximal inhibition concentration (IC50), values in parentheses: 97.5% confidence intervals.b Half maximal effective concentration (EC50), values in parentheses: 97.5% confidence intervals.c Data taken from ref. 49.d Data taken from ref. 32.e Data taken from ref. 47.f Data taken from ref. 50.g Data taken from ref. 51.h Data taken from ref. 34.i Not available.j Data taken from ref. 52.
1			5420 (4750–6230)	6230 (4980–8160)	1000 (777–1410)	9.99 (9.40–10.7)
2			3240 (2980–3530)	7470 (6860–8210)	>10 000	276 (265–287)
3			865 (782–957)	9670 (8700–11 000)	>10 000	242 (231–253)
4			13.3 (11.9–14.9)	9090 (8460–9800)	1960 (1370–2700)	59.7 (51.5–67.2)
5			533 (467–610)	614 (490–782)	>10 000	191 (172–210)
6			192 (169–217)	93.4 (82.3–105)	327 (226–484)	52.4 (47.4–57.2)
7			27.9 (24.0–32.5)	>10 000	>5000	171 (150–192)
8			28.2 (24.7–32.2)	>10 000	>5000	302 (274–324)
9			52.6 (49.5–55.8)	1680 (1550–1830)	>1000	323 (301–347)
10			37.0 (33.8–40.5)	533 (465–613)	>1000	123 (110–138)
11		R = ethyl	120 (110–130)^c	9900 (6000–23 000)^c	600 (530–690)^c	770 (720–810)^c
12		R = butyl	83.2 (74.1–93.3)^d	3580 (3030–4300)^e	177 (148–220)^e	84.7 (73.0–97.5)^f
13		R = hexyl	82.6 (75.4–90.5)^e	661 (535–843)^e	1.2^e	6.02 (2.00–16.2)^g
14		R = octyl	39.4 (36.7–42.3)^e	102 (91.2–115)^h	0.00175^e	0.047 (0.022–0.083)^g
15		R = decyl	12.3 (11.0–13.6)^e	21.9 (20–24)^h	0.000272	n.a.^i
					(0.000242–0.000308)^e
16		R = butyl	82.9 (74.8–91.9)^e	>20 000^j	2430 (1860–3320)^e	n.a.^i
17		R = hexyl	303 (278–330)^e	844 (770–933)^e	n.a.^i	n.a.^i
18		R = octyl	230 (211–251)^e	387 (339–446)^e	n.a.^i	n.a.^i

---

• the linkage chain length (ILs 1–4)

• the terminal alkyl side chain length (ILs 2, 5, 6)

• the linkage chain polarity (IL 4, 7, 9)

• the head group (IL 7, 8 and 9, 10) on the biodegradability and toxicity of the DILs.

The selection of halides and methylsulfonates enables the observed effects to be related solely to the cation moiety, owing to the low intrinsic toxicity of these anions.^27–29

For our preliminary hazard assessment of DILs we carried out different biodegradation tests, investigated the influence of DILs on a waste water treatment plant microbial community and performed toxicity tests at different levels of biological complexity, including enzyme inhibition tests (acetylcholinesterase), promyelocytic rat leukemia cells (IPC-81), limnic green algae (Scenedesmus vacuolatus) and water fleas (Daphnia magna). These test systems are well established and data for monocationic ILs as well as other reference compounds are available, which allows for a comparative hazard assessment.

This study is addressed to the users and developers of ILs, especially DILs, in all application and research fields in order to extend knowledge of their environmental properties and to create inherently safer products.

2 Experimental part

2.1 Chemicals

1H-Imidazole (>99.5%), 1-methylimidazole (>99.0%), 1-butylimidazole (>98.0%), 1-methylpyrrolidinium (>99.0%), tetraethylene glycol (TEG) (99%), hexaethylene glycol (HEG) (>97.0%), thionyl bromide, thionyl chloride (>99.0%), sodium azide (>99.0%), N,N-dimethylformamide (DMF) (>99.0%), N,N-diisopropylethylamine (DIPEA) (>95.0%), copper(I) bromide, diiodomethane (>99.0%), 1-bromohexane (>98.0%), 5,5′-dithio-bis-(2-nitrobenzoic acid) (DTNB), bovine serum albumin (BSA), K₂HPO₄, FeSO₄·7H₂O, (NH₄)₂Mo₇O₂₄·4H₂O, CaCl₂·2H₂O and urea were purchased from Sigma Aldrich (Steinheim, Germany); dibromoethane (>99.0%), dibromopropane (>98.0%) and dibromohexane (>98.0%) from Acros Organics (Geel, Belgium); phosphorus tribromide (>98.0%), propargyl alcohol (>99.0%), Na₂HPO₄, NaH₂PO₄, Na₂MoO₄·2H₂O, Titriplex III and 3,5-dichlorophenol from Merck (Darmstadt, Germany); methanesulfonyl chloride (>99.0%), acetylcholinesterase (from the electric eel), acetylthiocholine, NaCl, MgSO₄·7H₂O, H₃BO₃, ZnSO₄·7H₂O, FeCl₃·6H₂O, KNO₃, peptone from soybean and meat extracts from Fluka (Buchs, Switzerland); NaHCO₃ from GIBCO BRL (Eggenstein, Germany); KH₂PO₄, Ca(NO₃)₂·4H₂O and MnCl₂·4H₂O from Riedel-de Haën (Seelze, Germany); cell culture media, fetal bovine serum and phosphate buffer from Invitrogen Life Technologies (Frankfurt, Germany); WST-1 reagent from Roche Diagnostics GmbH (Mannheim, Germany).

2.2 Synthesis

The synthesis of 1 was performed as described in,²⁴ 2–6 were synthesized according to²³ and the preparation of 7–10 was previously described in.³⁰ Detailed information as well as NMR and MS- data are given in the ESI.† All DILs could be easily dissolved in water (>3 g L⁻¹). The miscibility with water was also described in other publications.^10,23,24

2.3 Biodegradation tests

2.3.1 Primary biodegradation. The primary biodegradation test was carried out using a modified version of OECD guideline 301 D.³¹ The concentration of the compound was monitored via ion chromatography (IC) for 28 days. Acquired from the domestic wastewater treatment plant at Delmenhorst (Germany), the inoculum was filtered and aerated before use. A mineral medium containing final concentrations of 8.5 g L⁻¹ KH₂PO₄, 21.75 mg L⁻¹ K₂HPO₄, 22.13 mg L⁻¹ Na₂HPO₄·2H₂O, 1.7 mg L⁻¹ NH₄Cl, 36.4 mg L⁻¹ CaCl₂·2H₂O, 22.5 mg L⁻¹ MgSO₄·7H₂O and 0.25 mg L⁻¹ FeCl₃ (pH 7.2) was added to the filtrate. In this test a bacteria number of 10⁶ cells L⁻¹ was applied (determined by Paddle-Tester; Hach Europe, Düsseldorf). Samples containing 200 μM (ca. 60–100 mg L⁻¹) of test substance were prepared, as well as blank samples (inoculated media without test substance), each in replicates. All samples were aerated at 20 °C in the dark during the test. Losses due to evaporation were checked regularly by weighing and balanced by the addition of water. The oxygen content was checked as well. For analysis of the biodegradation 2000 μL of all samples were taken at regular intervals and stored at −18 °C. At the end of the test period the samples were analyzed by IC (Metrohm 881 Compact IC system, equipped with online eluent degasser, 20 μL injection loop and a conductometric detector, maintained at 30 ± 0.1 °C, all Metrohm, Herisau, Switzerland). All chromatographic data were recorded by Metrohm software MagICNet version 1.1 compact. A silica-based (modified with carboxyl groups) Metrosep C4 ion exchange column (dimensions – 50 × 4.0 mm ID and 5 μm mean particle size) coupled with Metrosep C4 Guard and Metrosep RP Guard was used (all purchased from Metrohm, Herisau, Switzerland). An isocratic method using 40% acetonitrile in 6 mM nitric acid and a flow rate of 0.9 mL min⁻¹ was applied (limit of detection: 1–16 μM, limit of quantification: 4–50 μM). The percentage of biodegradation was determined from the ratio of the peak area to the initial concentration (day 0). Positive controls, imidazole and 1-methyl-3-octyl-imidazolium chloride, were tested to ensure the general activity of the inoculum.

2.3.2 Ready biodegradability according to OECD 301 F: manometric respirometry. The manometric respirometry test was performed according to OECD guideline 301 F.³¹ The biological oxygen demand of the substance was determined for 28 days using a BOD-System (OxiTop©, thermostatically controlled from WTW GmbH, Weilheim, Germany). Acquired from the wastewater treatment plant at Achim (Germany), the inoculum was filtered and aerated before use. A mineral medium containing final concentrations of 85 mg L⁻¹ KH₂PO₄, 217.5 mg L⁻¹ K₂HPO₄, 221.3 mg L⁻¹ Na₂HPO₄·2H₂O, 17 mg L⁻¹ NH₄Cl, 36.4 mg L⁻¹ CaCl₂·2H₂O, 22.5 mg L⁻¹ MgSO₄·7H₂O and 0.25 mg L⁻¹ FeCl₃ (pH 7.2) was added to the filtrate. An additional 1.16 mg L⁻¹ of allylthiourea was added to inhibit nitrification. The samples, containing inoculated media and ca. 120–150 mg L⁻¹ (giving 200 mg ThOD L⁻¹) substance, were prepared, as well as blank samples (inoculated media without test substance) and controls (inoculated media with benzoic acid). In this test a bacteria number of 10⁴ cells L⁻¹ was applied (determined by Paddle-Tester; Hach Europe, Düsseldorf). The flasks containing vessels with sodium hydroxide to ensure absorption of the carbon dioxide evolved were closed with gas-tight stoppers and stored in the dark at 20 °C. The oxygen consumption was determined manometrically. Biodegradation of the test substance was calculated by the oxygen uptake for the test substance (corrected by the oxygen demand of the blank samples) with respect to the theoretical oxygen demand (ThOD) of the substance and the amount of substance present in the sample. A positive control, benzoic acid, was tested to ensure the general activity of the inoculum.

2.4 Toxicity tests

2.4.1 Acetylcholinesterase inhibition assay. The inhibition of acetylcholinesterase (AchE) was measured using a colorimetric assay based on the reduction of the dye 5,5′-dithio-bis-(2-nitrobenzoic acid) (DTNB) by the enzymatically formed thiocholine moiety from the AchE substrate acetylthiocholine iodide. The assay is described in detail in Stock et al.³² Briefly, a dilution series of the test substances in phosphate buffer (0.02 M, pH 8.0) containing max. 1% methanol was prepared directly in the wells of a 96-well microtitre plate. DTNB (2 mM, 0.185 mg mL⁻¹ NaHCO₃ in phosphate buffer pH 8.0) and the enzyme (0.2 U mL⁻¹, 0.25 mg mL⁻¹ bovine serum albumin in phosphate buffer pH 8.0) were added to each well. The reaction was started by the addition of acetylthiocholine iodide (2 mM in phosphate buffer). The final test concentrations were 0.5 mM of DTNB and acetylthiocholine iodide and 0.05 U mL⁻¹ acetylcholinesterase, respectively. Each plate contained blanks (no enzyme) and controls (no toxicant). Enzyme kinetics were measured at 405 nm at 30 seconds intervals in a microplate-reader (MRX Dynatech) for 5 minutes. The enzyme activity was expressed as the slope of the optical density (in OD min⁻¹) from a linear regression. The relative toxicity of the samples was expressed as the percentage of enzymes activity compared to the controls.

2.4.2 Cell viability assay with IPC-81 cells. Cytotoxicity was determined for the promyelocytic leukemia rat cell line IPC-81.³³ Cultures of IPC-81 were grown in RPMI medium (with L-glutamine, without NaHCO₃, supplemented with 1% penicillin–streptomycin and 1% glutamine, pH 7) with 10% horse serum at 37 °C (5% CO₂).

The cytotoxicity assay was carried out according to Ranke et al.³⁴ Cell viability was measured using a colorimetric assay for 96 well plates with 2-(4-iodophenyl)-3-(4-nitrophenyl)-5-(2,4-disulfophenyl)-2H-tetrazolium monosodium salt (WST-1) reagent. Each plate contained blanks (no cells), controls (no toxicants), and substance in a 1 : 1 dilution series. Stocks of ILs were prepared in culture medium with 0.5% dimethylsulfoxide (DMSO) to improve the solubility of the substances. This DMSO concentration has been proven to be non-cytotoxic.³⁵ For the test, IPC-81 cells in a concentration of 15 × 10⁵ cells per mL⁻¹ (in RPMI with 8% fetal calf serum) were incubated for 44 h in 96-well plates in the presence of substance and for an additional 4 h in the presence of WST-1 reagent. Cell viability, measured as the ability to reduce WST-1, was observed photometrically at 450 nm in a microplate reader (MRX, Dynatech Laboratories, Chantilly, USA). The cytotoxicity of the compounds was expressed as a percentage of the cell viability measured as WST-1 reduction compared to controls. Each dose–response curve was recorded for at least 9 parallel dilution series on three different 96 well plates. Positive controls with carbendazim were checked at regular intervals.

2.4.3 Growth inhibition assay with the limnic green algae Scenedesmus vacuolatus. The reproduction inhibition assay with Scenedesmus vacuolatus was performed according to Matzke et al.³⁶ A synchronized culture of strain 211-15 (SAG (Culture Collection of Algae) Universität Göttingen, Göttingen) was used.³⁷ The stock culture was grown under photo-autotrophic conditions at 28 °C (±0.5 °C) in an inorganic, sterilized medium (pH 6.4) with saturating white light (intensity 22–33 klx) (Lumilux Daylight L 36 W-11 and Lumilux Interna L 36 W-41, Osram, Berlin, Germany). Cells were aerated with 1.5 vol% CO₂ and synchronized using a 14 h light and 10 h darkness cycle. The stock culture was diluted daily to a cell density of 5 × 10⁵ cells per mL.

For toxicity tests, autospores (young algal cells at the beginning of the growth cycle) were exposed to the test substances in different concentrations for one growth cycle (24 h). The tests were performed in sterilized glass tubes (20 mL Pyrex tubes sealed with caps containing a gas-tight Teflon membrane), the algae were stirred over the whole test period, and the test conditions were the same as for the stock culture, except the CO₂ source. Therefore, 150 μL of 0.15 M NaHCO₃ solution were added to each test tube. The initial and final cell number was measured using a Z2 Coulter Counter (Beckmann, Nürnberg, Germany). The relative toxicity of the samples was expressed as the percentage of growth compared to the controls. The tests were carried out at least three times for each substance with 2 replicates per concentration and a minimum of 6 controls (no toxicant).

2.4.4 Acute immobilization assay with Daphnia magna. The 48 h acute immobilization test with the crustacean Daphnia magna was assessed using the commercially available Daphtoxkit F (MicroBioTest Incorporation, Gent, Belgium) on the basis of the OECD 202 guideline.³⁸ The tests with neonates less than 24 h old, obtained from the hatching of ephippia, were performed at 20 °C in the dark. 5 pre-fed animals were incubated with the test substances in 10 mL of mineral medium (67.75 mg L⁻¹ NaHCO₃, 294 mg L⁻¹ CaCl₂·2H₂O, 123.25 mg L⁻¹ MgSO₄·7H₂O and 5.75 mg L⁻¹ KCl). The number of immobilized or dead organisms was checked after 24 and 48 h. The relative toxicity of the samples was expressed as the percentage of unaffected organisms compared to the controls. For each test 5 different concentrations of the ionic liquids in 5 replicates and 5 controls were investigated. All experiments were conducted at least twice. The sensitivity of the organisms to K₂Cr₂0₇ was checked routinely once a new batch of organisms was obtained.

2.4.5 Sludge inhibition test. The Activated Sludge–Respiration Inhibition Test was performed according to OECD guideline 209.³⁹ Acquired from the domestic wastewater treatment plant at Achim (Germany), the sludge was washed 3 times (decanted and resuspended with tap water) and continuously aerated. The dry mass of the sludge was measured gravimetrically and adjusted to 3 g L⁻¹ by the addition of tap water. For the toxicity test 4.8 mL synthetic feed composed of 16 g L⁻¹ peptone from soybeans, 11 g L⁻¹ meat extract, 3 g L⁻¹ urea, 0.7 g L⁻¹ NaCl, 0.4 g L⁻¹ CaCl₂·2H₂O, 0.2 g L⁻¹ MgSO₄·7H₂O and 3.67 g L⁻¹ K₂HPO₄·3H₂O, 75 mL activated sewage sludge and test substance in different concentrations were mixed, and water was added to give a total volume of 150 mL. The control samples were prepared identically, but no test substance was added. The mixtures was aerated and stirred. After 30 min the mixture was transferred to Karlsruher bottles and an oxygen electrode inserted in such a way that no headspace was produced over the liquid. The oxygen content was recorded at 30 s intervals for a maximum of 10 minutes or until the oxygen content fell below 2 mg L⁻¹. The respiration rate was expressed as the slope of the oxygen concentration from a linear regression. The relative toxicity of the samples was expressed as a percentage of the respiration rate compared to the controls. All experiments were conducted at least twice. The sensitivity of the sludge to 3,5-dichlorphenol was checked routinely once the sewage sludge had been obtained. The sludge was fed with 50 mL synthetic feed per liter at the end of each working day for a maximum of 4 days.

2.4.6 Effect data modeling. Dose–response curve parameters and plots were obtained using the drfit package (version 0.05–92) for the R language and environment for statistical computing (http://www.r-project.org).⁴⁰

3 Results and discussion

3.1 Biodegradation and sludge respiration inhibition

Primary biodegradation experiments were conducted in order to screen the biodegradability of all 10 DILs as well as imidazole and 1-methyl-3-octyl-imidazolium chloride (IM18 Cl) as the reference compounds. The initial concentration of the cation and the concentration after 28 days were determined by ion chromatography (IC) and used to calculate the percentage of primary degradation. The results are displayed in Table S1 (see ESI†). Primary degradation of imidazole and IM18 Cl was complete within 7 and 14 days, respectively, indicating the general biological activity of the inoculum. The tested DILs, however, remained stable within the measurement uncertainty levels and showed less than 5% primary degradation within 28 days. Thus, they remain recalcitrant towards microbial degradation under the applied “ready biodegradability” test conditions.

For DILs 4, 6 and 10 additional oxygen consumption tests with inoculum from a different domestic wastewater treatment plant (WWTP) were applied to enhance the variability of biological microorganisms. The compounds were chosen as the most appropriate candidates for biodegradation as they contain structural elements (a long alkyl or functionalized chain) that are known to increase the biodegradation potential.⁴¹ The results support the finding from the primary degradation study (Table S1†). Also in these experiments, less than 5% biodegradation was found for DILs 4 and 6. The 14% biodegradation observed for IL 10 can be attributed to the mineralization of the anionic moiety, which theoretically corresponds to 10% of the total oxygen demand of the DIL. Based on the low extent of biodegradation none of the structures can be classified as readily biodegradable, whereas the reference benzoic acid was completely mineralized within 3 days.

The results obtained in this study are in agreement with the observations made for monocationic ILs. For imidazolium-based monocationic ILs degradation of the side chain was observed if this consisted of eight or more carbon atoms.⁴¹ The longest substituted terminal alkyl side chain in our DILs was hexyl (6). Thus, the side chain might be too short for microbial degradation via ω-oxidation, which would explain the observed results. Although the set of DILs contained compounds with long spacer chains or ether links, this did not improve biodegradation of either the imidazolium- or the pyrrolidinium-based compounds. The linkage chain cannot be attacked via ω-oxidation and will only be accessible after head group removal. Tests with monocationic imidazolium-containing ILs had already shown that this core remains resistant to biodegradation.^41,42 For pyrrolidinium-based substances degradation was complete only when these were substituted with long or alcohol-functionalized side chains.⁴³ Therefore, the choice of longer terminal side chains (≥C8) or terminal hydroxyl groups, the introduction of ester groups in the linkage chain⁴⁴ and the use of other biodegradation-susceptible head groups such as pyridinium⁴⁵ might form DILs with enhanced biodegradability.

In the experiments discussed earlier, activated sludge from WWTPs was applied as inocula. If a DIL is toxic toward the microbial sludge community, the observed lack of biodegradation might be related to this. Therefore, a sludge inhibition test was performed and the O₂ respiration rate of the sludge was studied in the presence of a synthetic sewage feed at different DIL concentrations. The results, expressed as the half maximal inhibition concentration (IC₅₀), are shown in Table S1.† For IL 6 an IC₅₀ value of 380 μM (187 mg L⁻¹) was measured, which is higher than the concentration of the compound used in the tests (200 μM and 240 μM, respectively). However, residual inhibition of biodegradation cannot be excluded for this compound. No adverse effects on the microorganism's respiratory activity were found for any of the other compounds up to the highest tested concentration of 1 mM. Therefore, we can rule out inhibition of the microorganisms as a reason for the stability of these DILs. The experimentally found low toxicity of DILs to inocula is in good agreement with observations recently made for monocationic ILs substituted with short alkyl chains (≤-octyl).⁴⁶

3.2 Toxicity

Test systems of different complexity were chosen for a first estimate of the toxicity of the DILs. The results, expressed as the IC₅₀ or half maximal effective concentration (EC₅₀), are displayed in Table 1. Additional literature data of monocationic homologs are shown for comparison. We chose a series of 1-alkyl-3-methyl-imidazolium based and 1-alkyl-1-methyl-pyrrolidinium based ILs with different substituted alkyl side chains (-ethyl, -butyl, -hexyl, -octyl and -decyl) to act as reference compounds for lower (such as -ethyl) and higher (e.g. -decyl) acute IL toxicity.

3.2.1 Enzyme inhibition. The enzyme inhibition test with acetylcholinesterase (AchE) is an important biological marker in (eco)toxicology for evaluating the influence of chemicals on the central nervous system of organisms. To date, a large set of monocationic ILs have been investigated,^27,32,47 showing that a delocalized electron-deficient aromatic system as well as a certain lipophilicity are the key features defining their inhibitory potential.²⁷ In particular, increasing side chain length in 1-alkyl-3-methyl-imidazolium based ILs comes with a higher inhibitory potential (lower IC₅₀s), which lies between 12 and 120 μM (Table 1). The same trend was found for DILs, and the IC₅₀s span 2.5 orders of magnitude (13.3–5420 μM), depending on the length of the spacer (1–4). Whereas DILs with -methylene, -ethylene and -propylene spacers exhibit a significantly lower inhibition potential compared to monocationic ILs (1–3 vs. 11–15), the -hexylene spacer clearly reduces the IC₅₀, the extent of inhibition being similar to that found for 1-decyl-3-methyl-imidazolium chloride (15) (Fig. 1).

Fig. 1 Dose–response-curve for DILs with increasing linkage chain length and two monocationic ILs in acetylcholinesterase assay.

An enhanced inhibition was also found for DILs with increasing terminal side chain length (ILs 2, 5 and 6), but less pronounced (IC₅₀s within 1 decimal power) and even 2 terminally substituted -hexyl side chains (6) show a lower inhibition potential than short-chained monocationic compounds. Similar IC₅₀ values were measured for DILs composed of different head groups connected via polyether groups (7–10). The values in the range from 27.9 to 52.6 μM indicated an inhibition potential that is comparable to that of the monocationic IL IM18 Cl (14).

For AchE it is known that the active center is located in a narrow gorge of about 20 Å. The decamethonium cation has already been found to fit very well into this cleft.⁴⁸ When comparing the structures of decamethonium with DIL 4 the similarity is obvious. Thus, it can be assumed that this compound can be located in the same way, and prevents the substrate from reaching the enzymatic center, leading to the observed inhibition. The same might be true of DILs with similar or longer linkage chains (as for 7–10). On the other hand, the two head groups in the DILs 1–3, 5 and 6 might be too close to each other, and hence, the cation too bulky to fit into the gorge.

3.2.2 Cell toxicity with IPC-81 and Scenedesmus vacuolatus. In vitro testing with the leukemia rat cell line IPC-81 was used to screen for effects on basal cell functions and structures of cells. Basal cytotoxicity in mammalian cells has been reported to be similar in and therefore relevant to different organisms such as plants⁵³ and fish;⁵⁴ with restrictions, correlations to in vivo data have been found to be of predictive value.⁵⁵ The other cellular test in this study was performed with the limnic green algae Scenedesmus vacuolatus. Algae are primary producers and have a high relevance in the aquatic food chain. They are thus important test organisms for the hazard assessment of chemicals and for environmental legislation. The toxicity of monocationic ILs to IPC-81 and S. vacuolatus has been intensively screened.^{29,34,36,47,52} In general, a strong dependency between the hydrophobicity of the cation and the toxic effect was observed (Table 1; 11–15). Especially, S. vacuolatus is very sensitive to hydrophobic ILs. Simply by replacing the -ethyl side chain (11, EC₅₀ = 602 μM) with a -decyl one (15, EC₅₀ = 0.000272 μM) the EC₅₀ value was diminished by six orders of magnitude.²⁹ In contrast, most DILs caused comparably low toxicity towards algae and no EC₅₀ values up to the highest tested concentration of 1 to 10 mM could be determined. Only the most hydrophobic DIL with 2 terminally substituted -hexyl chains (6) caused stronger effects than IM12 Cl (11). The IPC-81 cell line was more sensitive to DILs than the algae, although the experimentally found EC₅₀s were often higher than the values determined for monocationic imidazolium-based ILs. The trend of greater toxicity with increasing cation hydrophobicity was not observed with increasing spacer length (1–4), but did manifest itself when the length of the terminal side chains was increased (2, 5, 6). The comparable lower EC₅₀ value of IL 1 might be explained by the presence of iodide, tested as the potassium salt (EC₅₀ 4998 μM). Iodide is known to cause apoptosis and necrosis in the rat thyroid cell line (FTLR-5) at high doses.⁵⁶ DILs 7 and 8, which contain a tetraethylene glycol linkage chain, showed no adverse effect up to 10 mM. By elongating the spacer to hexaethylene glycol and introducing a triazole group (ILs 9 and 10), the half maximal effect concentration is reduced. As in the acetylcholinesterase test, the imidazolium-based IL 9 (EC₅₀ 1690 μM) had a lower cytotoxic effect than the pyrrolidinium IL 10 (EC₅₀ 533 μM). This is contrary to the findings for monocationic ILs (12–14 vs. 16–18). However, this trend is not consistent since DIL 7 and 8 showed the expected tendency. Further testing is necessary to explain these results. Generally in this test, the cytotoxicity of the DILs is in the same range as that for the most frequently investigated monocationic ILs.

3.2.3 Immobilization toxicity with Daphnia magna. The crustacean Daphnia magna plays a very important ecological role in freshwater habitats, because it is a very efficient filter feeder and serves as a major food source for a whole range of aquatic invertebrates and vertebrates. It is a sensitive test organism often used for assessing the hazards posed by toxicants.

The sensitivity of D. magna to DILs is higher compared with the test using IPC-81 and algae. Again, IL 1 appears to be the most toxic DIL (EC₅₀ = 10 μM), but tests with KI (EC₅₀ = 17.6 μM) indicates once more the influence of iodide on the toxicity of 1. For the other compounds the EC₅₀s are observed to be dependent on the terminal alkyl chain and the linkage chain length (IL 2–6). The introduction of a polyethylene glycol-containing spacer reduces the toxicity (7–10). In comparison to monocationic ILs the EC₅₀s of DILs generally lay between 11 (-ethyl) and 12 (-butyl).

4 Conclusion

An initial evaluation of the hazard potential of 10 synthesized dicationic cations is presented. In comparison with standard dialkylimidazolium ILs, the levels of enzyme inhibition towards acetylcholinesterase and the acute toxicity towards IPC-81, S. vacuolatus and D. magna lie in many cases below the toxicity observed for monocationic ILs. Especially the use of short terminal side chains and head groups connected via polyethylene glycol or short alkyl chains could be identified as structural elements reducing the toxicity. The high thermal and electrochemical stability of DILs is reflected in their stability towards biodegradation processes. None of the DILs could be degraded during the biodegradation experiments. However, bearing in mind what is known about monocationic ILs it might be possible to design new DILs with an enhanced biodegradation potential. In particular, the use of other head groups such as pyridinium, known to be susceptible to biodegradation,⁴⁵ appears promising. From a structural point of view multivalent ILs certainly have considerable potential that fulfill technical needs and pose a reduced hazard to humans and the environment.

Acknowledgements

The authors gratefully thank the whole MINILUBES and UFT team for their advice and support. Thanks are due to Prof. Wolfgang Binder and his research group for synthesizing and providing bis-1,11-[(3-methyl-1H-imidazolium-1-yl)]-(3,6,9-trioxaundecane) dichloride and bis-1,11-[(1-methyl-pyrrolidinium-1-yl)]-(3,6,9-trioxaundecane) dichloride. The authors would like to acknowledge the financial support of the European project MINILUBES (FP7 Marie Curie ITN network 216011-2) by the European Commission.

Notes and references

1. T. Welton, Chem. Rev., 1999, 99, 2071–2084.
2. F. Shi, Y. Gu, Q. Zhang and Y. Deng, Catal. Surv. Asia, 2004, 8, 179–186.
3. S. Seki, Y. Kobayashi, H. Miyashiro, Y. Ohno, A. Usami, Y. Mita, N. Kihira, M. Watanabe and N. Terada, J. Phys. Chem. B, 2006, 110, 10228–10230.
4. R. Kawano, H. Matsui, C. Matsuyama, A. Sato, M. A. B. H. Susan, N. Tanabe and M. Watanabe, J. Photochem. Photobiol., A, 2004, 164, 87–92.
5. N. V. Plechkova and K. R. Seddon, Chem. Soc. Rev., 2008, 37, 123–150.
6. J. Ranke, S. Stolte, R. Störmann, J. Arning and B. Jastorff, Chem. Rev., 2007, 107, 2183–2206.
7. M. Petkovic, K. R. Seddon, L. P. N. Rebelo and C. Silva Pereira, Chem. Soc. Rev., 2011, 40, 1383–1403.
8. T. P. Thuy Pham, C.-W. Cho and Y.-S. Yun, Water Res., 2010, 44, 352–372.
9. S. Stolte, S. Steudte, A. Igartua and P. Stepnowski, Curr. Org. Chem., 2011, 15, 1946–1973.
10. J. L. Anderson, R. Ding, A. Ellern and D. W. Armstrong, J. Am. Chem. Soc., 2005, 127, 593–604.
11. R. P. Singh and J. M. Shreeve, Chem. Commun., 2003, 1366–1367.
12. H. Shirota, T. Mandai, H. Fukazawa and T. Kato, J. Chem. Eng. Data, 2011, 56, 2453–2459.
13. Y.-S. Ding, M. Zha, J. Zhang and S.-S. Wang, Colloids Surf., A, 2007, 298, 201–205.
14. C.-M. Jin, C. Ye, B. S. Phillips, J. S. Zabinski, X. Liu, W. Liu and J. M. Shreeve, J. Mater. Chem., 2006, 16, 1529–1535.
15. F. Pagano, C. Gabler, P. Zare, M. Mahrova, N. Dörr, R. Bayon, X. Fernandez, W. H. Binder, M. Hernaiz, E. Tojo and A. Igartua, Proc. Inst. Mech. Eng., Part J, 2012, 226, 952–964.
16. G. Yu, S. Yan, F. Zhou, X. Liu, W. Liu and Y. Liang, Tribol. Lett., 2007, 25, 197–205.
17. J. L. Anderson and D. W. Armstrong, Anal. Chem., 2005, 77, 6453–6462.
18. K. Huang, X. Han, X. Zhang and D. Armstrong, Anal. Bioanal. Chem., 2007, 389, 2265–2275.
19. Z. Zhang, H. Zhou, L. Yang, K. Tachibana, K. Kamijima and J. Xu, Electrochim. Acta, 2008, 53, 4833–4838.
20. J. Y. Kim, T. H. Kim, D. Y. Kim, N.-G. Park and K.-D. Ahn, J. Power Sources, 2008, 175, 692–697.
21. C. Zafer, K. Ocakoglu, C. Ozsoy and S. Icli, Electrochim. Acta, 2009, 54, 5709–5714.
22. J.-C. Chang, W.-Y. Ho, I.-W. Sun, Y.-L. Tung, M.-C. Tsui, T.-Y. Wu and S.-S. Liang, Tetrahedron, 2010, 66, 6150–6155.
23. M. Lee, Z. Niu, C. Slebodnick and H. W. Gibson, J. Phys. Chem. B, 2010, 114, 7312–7319.
24. Q. Liu, F. van Rantwijk and R. A. Sheldon, J. Chem. Technol. Biotechnol., 2006, 81, 401–405.
25. M. Messali, Z. Moussa, A. Y. Alzahrani, M. Y. El-Naggar, A. S. ElDouhaibi, Z. M. a. Judeh and B. Hammouti, Chemosphere, 2013, 91, 1627–1634.
26. L. Ford, J. R. Harjani, F. Atefi, M. T. Garcia, R. D. Singer and P. J. Scammells, Green Chem., 2010, 12, 1783–1789.
27. J. Arning, S. Stolte, A. Böschen, F. Stock, W.-R. Pitner, U. Welz-Biermann, B. Jastorff and J. Ranke, Green Chem., 2008, 10, 47–58.
28. S. Stolte, J. Arning, U. Bottin-Weber, M. Matzke, F. Stock, K. Thiele, M. Uerdingen, U. Welz-Biermann, B. Jastorff and J. Ranke, Green Chem., 2006, 8, 621–629.
29. S. Stolte, M. Matzke, J. Arning, A. Böschen, W.-R. Pitner, U. Welz-Biermann, B. Jastorff and J. Ranke, Green Chem., 2007, 9, 1170–1179.
30. P. Zare, M. Mahrova, E. Tojo, A. Stojanovic and W. H. Binder, J. Polym. Sci., Part A: Polym. Chem., 2013, 51, 190–202.
31. Test no. 301: Ready Biodegradability, OECD Guidel. Test. Chem., 1992, no. 301, 1–62.
32. F. Stock, J. Hoffmann, J. Ranke, R. Störmann, B. Ondruschka and B. Jastorff, Green Chem., 2004, 6, 286–290.
33. N. Lacaze, G. Gombaud-Saintonge and M. Lanotte, Leuk. Res., 1983, 7, 145–154.
34. J. Ranke, K. Mölter, F. Stock, U. Bottin-Weber, J. Poczobutt, J. Hoffmann, B. Ondruschka, J. Filser and B. Jastorff, Ecotoxicol. Environ. Saf., 2004, 58, 396–404.
35. F. Stock, Dissertation, University of Bremen, 2004.
36. M. Matzke, S. Stolte, K. Thiele, T. Juffernholz, J. Arning, J. Ranke, U. Welz-Biermann and B. Jastorff, Green Chem., 2007, 9, 1198–1207.
37. M. Faust, R. Altenburger, W. Bodeker and L. H. Grimme, Schriftenr. Ver. Wasser-, Boden- Lufthyg., 1992, 89, 311–321.
38. Test no. 202: Daphnia sp. Acute Immobilisation Test, OECD Guidel. Test. Chem., 2004, 1–12.
39. Test no. 209: Activated Sludge Respiration Inhibition Test (Carbon and Ammonium Oxidation), OECD Guidel. Test. Chem., 2010, 1–18.
40. R Core Team, R Foundation for Statistical Computing, Version 2.15.1, 2012.
41. S. Stolte, S. Abdulkarim, J. Arning, A.-K. Blomeyer-Nienstedt, U. Bottin-Weber, M. Matzke, J. Ranke, B. Jastorff and J. Thöming, Green Chem., 2008, 10, 214–224.
42. E. Rorije, F. Germa and B. Philipp, SAR QSAR Environ Res, 2002, 13, 199–204.
43. J. Neumann, S. Steudte, C.-W. Cho, J. Thöming and S. Stolte, Green Chem., 2013 , submitted.
44. N. Gathergood, P. J. Scammells and M. T. Garcia, Green Chem., 2006, 8, 156–160.
45. K. M. Docherty, J. K. Dixon and C. F. Kulpa, Biodegradation, 2007, 18, 481–493.
46. M. Markiewicz, M. Piszora, N. Caicedo, C. Jungnickel and S. Stolte, Water Res., 2013, 47, 2921–2928.
47. http://www.il-eco.uft.uni-bremen.de.
48. P. H. Axelsen, M. Harel, I. Silman and J. L. Sussman, Protein Sci., 1994, 3, 188–197.
49. S. Steudte, P. Stepnowski, C.-W. Cho, J. Thöming and S. Stolte, Chem. Commun., 2012, 48, 9382–9384.
50. R. J. Bernot, M. A. Brueseke, M. A. Evans-White and G. A. Lamberti, Environ. Toxicol. Chem., 2005, 24, 87–92.
51. D. J. Couling, R. J. Bernot, K. M. Docherty, J. K. Dixon and E. J. Maginn, Green Chem., 2006, 8, 82–90.
52. S. Stolte, J. Arning, U. Bottin-Weber, A. Muller, W.-R. Pitner, U. Welz-Biermann, B. Jastorff and J. Ranke, Green Chem., 2007, 9, 760.
53. U. Kristen, Toxicol. In Vitro, 1997, 11, 181–191.
54. A. Castaño and M. J. Gómez-Lechón, Toxicol. In Vitro, 2005, 19, 695–705.
55. M. J. Garle, J. H. Fentem and J. R. Fry, Toxicol. In Vitro, 1994, 8, 1303–1312.
56. J. Golstein and J. E. Dumont, J. Endocrinol. Invest., 1996, 19, 119–126.

---

Footnote

† Electronic supplementary information (ESI) available: See DOI: 10.1039/c3ra45675g

---