Ionic liquids based microwave-assisted extraction of lichen compounds with quantitative spectrophotodensitometry analysis

Ionic liquids based microwave-assisted extraction of lichen compounds with quantitative spectrophotodensitometry analysis

Analytica Chimica Acta 707 (2011) 69–75 Contents lists available at SciVerse ScienceDirect Analytica Chimica Acta journal homepage: www.elsevier.com...

855KB Sizes 6 Downloads 91 Views

Analytica Chimica Acta 707 (2011) 69–75

Contents lists available at SciVerse ScienceDirect

Analytica Chimica Acta journal homepage: www.elsevier.com/locate/aca

Ionic liquids based microwave-assisted extraction of lichen compounds with quantitative spectrophotodensitometry analysis Sarah Bonny a , Ludovic Paquin b , Daniel Carrié b , Joël Boustie a , Sophie Tomasi a,∗ a Equipe PNSCM “Produits naturels – Synthèses – Chimie Médicinale”, UMR CNRS 6226 Sciences chimiques de Rennes, UFR Sciences Pharmaceutiques et Biologiques, Univ. Rennes 1, Université Européenne de Bretagne, 2 Avenue du Pr. Léon Bernard, F-35043 Rennes, France b Equipe ICMV “Ingénierie Chimique et Molécules pour le Vivant”, UMR CNRS 6226, Univ. Rennes 1, Université Européenne de Bretagne, 263 Avenue du General Leclerc, F-35042 Rennes, France

a r t i c l e

i n f o

Article history: Received 10 June 2011 Received in revised form 26 August 2011 Accepted 12 September 2011 Available online 16 September 2011 Keywords: Pertusaria pseudocorallina Lichens, Polyphenolic compounds Ionic liquids Microwave-assisted extraction (MAE) Heat-extraction (HE)

a b s t r a c t Ionic liquids based extraction method has been applied to the effective extraction of norstictic acid, a common depsidone isolated from Pertusaria pseudocorallina, a crustose lichen. Five 1-alkyl-3methylimidazolium ionic liquids (ILs) differing in composition of alkyl chain and anion were investigated for extraction efficiency. The extraction amount of norstictic acid was determined after recovery on HPTLC with a spectrophotodensitometer. The proposed approaches (IL-MAE and IL-heat extraction (IL-HE)) have been evaluated in comparison with usual solvents such as tetrahydrofuran in heat-reflux extraction and microwave-assisted extraction (MAE). The results indicated that both the characteristics of the alkyl chain and anion influenced the extraction of polyphenolic compounds. The sulfate-based ILs [C1 mim][MSO4 ] and [C2 mim][ESO4 ] presented the best extraction efficiency of norstictic acid. The reduction of the extraction times between HE and MAE (2 h–5 min) and a non-negligible ratio of norstictic acid in total extract (28%) supports the suitability of the proposed method. This approach was successfully applied to obtain additional compounds from other crustose lichens (Pertusaria amara and Ochrolechia parella). © 2011 Elsevier B.V. All rights reserved.

1. Introduction Room-temperature ionic liquids (ILs) are salts, with melting temperature below the boiling point of water, resulting from combinations of bulky organic cations and inorganic or large organic anions [1]. These neoteric solvents are a class of novel green chemicals which are designed to replace traditional volatile organic solvents in industrial processes. The potential of ILs in both academic and industrial fields is due to their negligible vapor pressure, good thermal sensibility, low or virtually no volatility, good miscibility with water and organic solvents, as well as extractability for various organic compounds [2–4]. ILs are regarded as potentially environmentally benign solvents although recent publications indicated that the toxicity of imidazolium based ILs increases with the alkyl chain length. Such compounds have been shown to be either more toxic to freshwater algae [5] or less toxic to watercress [6] than the conventional solvents. Additionally, microwave-assisted extraction (MAE) has received increasing attention as an alternative to classical solid–liquid extraction methods since traditional extraction techniques require long extraction times. Considered that ILs

∗ Corresponding author. Tel.: +33 0 2 23 23 48 17; fax: +33 0 2 23 23 47 04. E-mail address: [email protected] (S. Tomasi). 0003-2670/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.aca.2011.09.009

can efficiently absorb microwave energy, it is rather an interesting challenge to use ILs as solvent for the MAE of various biomolecules from solid samples. 1-Alkyl-3-methylimidazolium based ionic liquids aqueous solutions were investigated as solvents for the extraction of polyphenolic compounds from Psidium guajava leaves and Smilax china tubers [7], of lignans from Schisandra chinensis [8] and of alkaloids from medicinal plants [9], indicating that ILs had potential applications in the MAE of bioactive substances from various sources. Recently, protic ionic liquids (N,N-dimethyl-N-(2-hydroxyethoxyethyl)ammonium propionate (DMHEEAP) and N,N-dimethyl(cyanoethyl)ammonium propionate (DMCEAP)) were used as solvents in MAE to extract lactones from Ligusticum chuanxiong [10]. While the extraction mechanism of MAE using IL was similar to traditional organic solvents, the efficiency of MAE was found much higher than standard extraction. Lichens are fascinating organisms, known as a stable, selfsupporting symbiont involving algae or cyanobacteria and fungi [11]. They produce characteristic secondary metabolites which are unique and allow them to survive under various harsh environmental conditions (e.g., at high altitude, direct sunlight). These compounds have various biological activities such as: antiviral, antitumor, anti-inflammatory, antioxidant properties [12,13]. Among them, depsides, depsidones and depsones are the major representative groups. However they are often extracted in low

70

S. Bonny et al. / Analytica Chimica Acta 707 (2011) 69–75 CH 3

O

O

H3C

O

CH3

O

H3 C

OH

O O

OH

O O

O

O O HO

R

HO

OH

R = CHO : norstictic acid R = CH2OH : connorstictic acid

C 5H11

O

O

HO

O

OH

CH 3

R

O O

O

OH

O

O O

HO HO α -alectoronic acid

O

lecanoric acid

variolaric acid

OH

OH

OH

O

OCH3 R

C 5H11

COOH

R = C 5 H 11 :picrolichenic acid R = C 5 H 11 or C 3H 7 : picrolichenic acid derivatives

Fig. 1. Chemical structures: norstictic acid; connorstictic acid; variolaric acid; lecanoric acid; ␣-alectoronic acid; picrolichenic acid and derivatives.

quantities due to a reduced availability of the material and their limited solubility in organic solvents. Pertusaria pseudocorallina is a crustose lichen collected on seashore rocks and described to contain as major compounds, two depsidones corresponding to norstictic acid and its methyl alcoholic derivative connorstictic acid [14] (Fig. 1). Norstictic acid has already demonstrated significant antioxidant activities [15]. These polyphenolic compounds are usually insoluble in water and are extracted with volatile organic solvents. Previously we demonstrated their stability under MAE at temperatures even up to 100 ◦ C for 15 min without degradation using tetrahydrofuran or acetone [16]. In this work, ILs were investigated as alternative solvents in MAE for the extraction of norstictic acid with regard to the balance of their extraction yields, extraction time and selectivity. The ionic liquids based microwave-assisted extraction (IL-MAE) performed using five 1-alkyl-3-methylimidazolium ionic liquids were compared to classical extraction approaches. The efficiency of the methods was evaluated by quantification after recovery of norstictic acid in the extracts from P. pseudocorallina using HPTLC coupled to a Camag® spectrophotodensitometer. This approach was then extended to other crustose lichens Pertusaria amara [17] and Ochrolechia parella [18] for the extraction of picrolichenic acid along with derivatives and variolaric, lecanoric and ␣-alectoronic acids, respectively. The structures of these isolated metabolites are reported in Fig. 1. 2. Experimental 2.1. Plant materials and reagents P. pseudocorallina (Sw.) Arn. (Pertusariaceae) was collected on seashore rocks near Avranches, France, in September 2008. Voucher specimens have been deposited at the herbarium of Pharmacognosy and Mycology, Rennes, France (REN-ABB) with the reference number JB/07/e9. After cleaning and identification, lichens were air-dried, ground to fine powder just before extraction and stored at room temperature in a closed glass. Acetone, tetrahydrofuran, isopropanol were purchased from Carlo Erba® (Milano, Italy); toluene and diethyl ether and formic acid from Prolabo® . 1-Methylimidazole, diethylsulfate, dimethylsulfate, chlorobutane, bis((trifluoromethyl)sulfonyl)imide lithium (Fluka) were used as received. Norstictic acid standard of 98% purity used in quantitative analysis was isolated from P. pseudocorallina after purification procedure using flash chromatography with cyclohexane/diethyl ether/tetrahydrofuran/formic acid (50/40/50/2) as a mobile phase

and a 15 mL min−1 flow rate. Norstictic acid was confirmed by comparison of its 1 H NMR and 13 C NMR spectra with previous data [17]. HREIMS in a negative ion mode confirmed the molecular formula of norstictic acid of C18 H12 O9 with a typical molecular ion cluster at m/z 371.04086 (calcd 371.0408). Picrolichenic acids (PA and PA derivatives) and variolaric (VA), lecanoric (LA), ␣-alectoronic acids were obtained with a 98% purity after extraction with THF from P. amara [17] and with acetone from O. parella [18], respectively and successive precipitation and column chromatography on silica gel. 2.2. Apparatus A CEM Explorer® 24, a focused monomode microwaves apparatus, was used for the experiments. The apparatus was run under pressure in closed system with programmable temperature, heating power, and irradiation time. A Camag® automatic sampler III (ATS3) was used for sample application in TLC. The samples (2 mL) and standard solutions (5 mL) were spotted in the form of bands of 4 mm width, positioned at 15 mm from the bottom of TLC plates (Merck silica gel 60F254, 20 cm × 10 cm), with a Camag® microliter syringe. The mobile phase was toluene/ethyl acetate/formic acid (70/25/5). Camag® TLC Scanner II was used for densitometric evaluation of thin-layer chromatograms and quantification was carried out under UV light at 270 nm (deuterium lamp) in absorbance/reflectance mode. The HPLC system equipped with a Kontron 325 pump and a diode array detector (DAD) was used for some measurements. The injector is fitted with a 20 ␮L external loop. A C18 Spherisorb column (250 × 4.6 mm, 5 ␮m) was used as analytical column. 1 H and 13 C NMR spectra were performed using a Bruker Avance III 300 NMR spectrometer. 2.3. Synthesis of ILs The structures of ILs used here are reported in Table 1. 1,3-Dimethylimidazolium methylsulfate [C1 mim][MSO4 ] and 1ethyl-3-methylimidazolium ethylsulfate [C2 mim][ESO4 ] were synthesized according to the literature procedure [19]. Another procedure was used for the bis{(trifloromethyl)sulfonyl}imide basedionic liquids ([C2 mimOH][NTf2 ], [C3 mim][NTf2 ], [C4 mim][NTf2 ] [20,21]. Briefly, the synthesis of [C2 mimOH][Cl], [C3 mim][Cl] and [C4 mim][Cl] were realized under microwave irradiations focused at 120 ◦ C (150 ◦ C for [C4 mim][Cl]) during 30 min. The NTf2 based ILs were prepared by reaction of the three corresponding chlorides and LiNTf2 at a molar ratio 1:1.

S. Bonny et al. / Analytica Chimica Acta 707 (2011) 69–75

71

Table 1 Chemical structures and properties of ionic liquids. Ionic liquids

Cations

Anions

MW

Form (25 ◦ C)

Water miscibility

Water content %

[C1 mim][MSO4 ]

N

N

CH3 SO4 −

160

Solid

Totally miscible

1.20

[C2 mim][ESO4 ]

N

N

C2 H5 SO4 −

188

Liquid

Totally miscible

1.62

[C2 mimOH][NTf2 ]

N

N

NTf2 −

454

Liquid

Non miscible

1.06

[C3 mim][NTf2 ]

N

N

NTf2 −

466

Liquid

Non miscible

1.09

[C4 mim][NTf2 ]

N

N

NTf2 −

480

Liquid

Non miscible

1.37

OH

These ILs were dried at 70 ◦ C under vacuum during several hours and their purity was then checked by 1 H and 13 C NMR spectra at room temperature in D2 O or CDCl3 . NMR spectra are identical to those previously described [22] and indicated their good purity (>95%). The water content was determined using Karl Fischer titration (Table 1). 2.4. MAE procedure The IL-MAE procedure was performed using CEM Explorer® 24. P. pseudocorallina, P. amara or O. parella (0.25 g) were mixed with 5 mL of each ionic liquid, and then the suspensions were irradiated under microwave heating for 5 min at 100 ◦ C. Regular MAE was performed using 5 mL of THF as solvent for 5 min at 100 ◦ C. All extraction experiments were repeated three times. After the irradiation, the obtained extracts were filtered through cotton to remove lichen residue and were cooled down to room temperature. 2.5. Conventional extraction method 2.5.1. Organic solvents Heat-reflux extraction (HE) was chosen as the reference process for extraction of norstictic acid. 0.5 g sample was mixed with 10 mL of THF in a round-bottom flask and the mixture was boiled at 70 ◦ C for 2 h under reflux. The extraction matrix was re-extracted two times using solvent to perform the total extraction of norstictic acid. The suspensions obtained after extraction were filtered through a paper and the filtrate evaporated to dryness. The final extracts were diluted with identical solvent to 1 mg mL−1 concentration. 2.5.2. ILs 0.25 g samples (P. pseudocorallina) were extracted with 5 mL of different IL solutions, heated in an oil bath for 2 h at 100 ◦ C. The resulted extracts were filtered through cotton, and then the filtrate cooled down at room temperature. 2.6. HPLC analysis The extracts diluted in THF at 5 mg ml−1 were filtered before injection. The mobile phase consisted of solvent A (0.6% acetic acid aqueous solution) and solvent B (MeOH) (60/40) with a flow rate fixed at 0.8 mL min−1 . The UV detection wavelengths were 270 nm and each injection volume was 20 ␮L. The column temperature was ambient. Norstictic and connorstictic acids identification was

carried out by comparing their retention time with corresponding peaks in the standard solution.

2.7. Recovery of compounds and quantification 2.7.1. Recovery The proposed process involves liquid–liquid extraction with water and diethyl ether followed by separation–precipitation with a mixture of acetone and water (1:1, v/v) (Fig. 2). After the extraction process, the raw extracts were filtered with cotton and the filtrate was kept at room temperature to allow the precipitation of compounds. The precipitate obtained (precipitate 1) was separated from the filtrate then water was added as co-extractant (2:3, v/v) and diethyl ether as extractant (0.5 mL × 2 mL). The whole was stirred during 15 min then put to centrifugation at 4000 tr min−1 during 15 min. The superior organic phase which contains some metabolites (terpenoids as apolar metabolites, depsidones. . .) was evaporated. A mixture of acetone/water (1:1, v/v) was added to the aqueous phase to precipitate targeted metabolites (precipitate 2). Both precipitates are mixed, washed with acetone then dried under reduced pressure (Fig. 2). The final extracts and the dried organic extract were prepared in tetrahydrofuran, corresponding to 0.5 mg mL−1 crude extract and were analyzed by HPTLC coupled to a spectrophotodensitometer and by HPLC.

2.7.2. Quantification using Camag® spectrophotodensitometer-method validation Standard solutions for each compound (0.05–0.3 mg mL−1 in THF) were used for the preparation of a 5-point calibration curve corresponding to an amount of 0.2–1.2 ␮g. The 0.5 mg mL−1 stock solutions in THF are quantified according to the procedure already described [16]. The method validation of the chromatographic determination with TLC scanner II was studied. A series of standard solutions with the concentration range 0.01–10 mg mL−1 was prepared to determine the linearity of this method. The linear range was from 0.02 to 0.5 mg mL−1 . The correlation coefficient of the calibration curve was 0.9989. The limit of detection and the limit of quantification were 0.01 and 0.02 mg mL−1 , respectively. The repeatability of the chromatographic determination with TLC scanner II, expressed in RSDr, was also considered in order to determine the experimental errors and evaluated as 3.2%.

72

S. Bonny et al. / Analytica Chimica Acta 707 (2011) 69–75

Fig. 2. Process of IL-based extraction from P. pseudocorallina, recovery of depsidones (NA) and possible regeneration of ILs.

The extraction efficiency of the compounds in samples was defined by the determination of the extraction yields as follows: yield (%) =

mass of the compound in extract × 100 mass of the material

The ratio in the total yield of the compounds was defined as follows: ratio in the total yield (%) =

mass of the compound in extract × 100 sum of mass of the compound of the extract

The total mass of norstictic acid in sample was determined by analysis of the total extraction. 3. Results and discussion 3.1. Choice of ILs Preliminary tests were performed using 18 ILs corresponding to combinations of 1-alkyl-3-methylimidazolium alkylsulfate, 1-alkyloxy-3-methylimidazolium, ammonium and phosphonium based ionic liquids with chloride, hexafluorophosphate and bis(trifluoromethylsulfonyl)imide as anions. The solid–liquid ratio (1:20) was kept the same for all experiments and determined from previous studies [16]. A 2-fold increase in the solid amount decreased the extraction efficiency due to uneasy filtration. Regarding their TLC profiles, [C1 mim][MSO4 ], [C2 mim][ESO4 ], [C4 mim][PF6 ], Aliquat 336, [PBu4 ][Cl], [PBu4 ][PF6 ], [C3 mim][NTf2 ], [C3 mimOH][NTf2 ] and [C6 mimOH][NTf2 ] appeared to be the best suitable ILs for the extraction of the tested lichens. Indeed, some of them allowed selective extraction of norstictic acid ([C2 mim][ESO4 ] and [C3 mimOH][NTf2 ]) or along with other compounds ([C1 mim][MSO4 ]). Considering the possibility of ILs regeneration, their easy synthesis and aromatic feature able to interact with ␲-systems [23,24] of the required lichen compounds, imidazolium cations were selected to extract compounds from P. pseudocorallina in order to evaluate their performance in HE and MAE approaches. Thus, the most interesting ILs according to the given objectives remain [C1 mim][MSO4 ], [C2 mim][ESO4 ] and [C3 mim][NTf2 ] because besides extracting the targeted substances, most of them allowed to get back them and to envisage their quantification. Moreover, we decided to use [C4 mim][NTf2 ] and [C2 mimOH][NTf2 ]

to evaluate the influence of the size of the alkyl chain along with the presence of a hydroxyl group for the extraction efficiency. 3.2. Effect of IL structures Previous studies have demonstrated the influence of the structure of ILs on their physicochemical properties [25,26], which affect the extraction efficiency of target analytes. The effects by change in anions and aliphatic chain length on the extraction yields were thus investigated in this work. In the series of ILs studied here, the anion identity greatly influences the water miscibility [27,28]. Thus, the NTf2 anion was tested with three imidazolium cations as well as sulfate anions (Table 1). The sulfate based-ILs presented the best norstictic acid extraction yield with similar yields for [C1 mim][MSO4 ] using MAE (3.2%) and [C2 mim][ESO4 ] under HE (3.8%) (Fig. 3). The extraction process used influenced the recovery of norstictic acid from crude extract

Fig. 3. Effect of ILs on the extraction efficiency of norstictic acid from P. pseudocorallina. Sample: 0.25 g, extractant volume: 5 mL, extraction time: 5 min (MAE)–2 h (HE). No result was obtained with [C4 mim][NTf2].

S. Bonny et al. / Analytica Chimica Acta 707 (2011) 69–75

of [C2 mim][ESO4 ] as indicated by the efficiency decrease between HE (with 3.8% yield) and MAE (0.5% yield) (Fig. 3). The low extraction efficiency under microwave irradiations could be explained by the high ability of this IL to absorb and transfer microwave energy leading to the degradation of lichen sample. It is for this reason that the temperature for IL-MAE was limited to 100 ◦ C and was applied to IL-HE. As shown in Fig. 3, the extraction yield was also decreased from RSO4 − to NTf2 − due to the decrease in the water miscibility (Table 1). In the case of NTf2 -ILs, the increase in alkyl chain from propyl to butyl dramatically decreased the extraction efficiency (Fig. 3). Thus, [C3 mim][NTf2 ] led to extraction of norstictic acid (1.6%) whereas the extraction yields of [C4 mim][NTf2 ] were not measurable. These results suggested that alkyl chain length has influence on the extraction and the recovery of norstictic acid and could be related to the hydrophobicity which increased with the length of the chain (Table 1). The substitution of a terminal methyl group by a hydroxyl group did not however improve the extraction yield (Fig. 3). Similar extraction yields for each extraction method used were thus obtained between [C2 mimOH][NTf2 ] and [C3 mim][NTf2 ]. A 0.5% extraction yield was performed under HE for these two ILs while this efficiency was 3-fold higher using MAE (around 1.5%) (Fig. 3), suggesting their ability to heat up to a good extent under microwaves effect.

3.3. Comparison of the proposed IL-MAE procedure with classical methods In order to compare the extraction efficiency using ILs with organic solvents in MAE and HE processes, tetrahydrofuranwas

73

used to extract norstictic acid from P. pseudocorallina. We have already demonstrated that THF led to the best extraction of this sample specially under MAE [16] and this solvent avoids the degradation of the lichen samples sensitive to hydrolysis by aqueous or polar solvent such as MeOH [29]. The major constituents of heatreflux and MAE extracts have been characterized using HPLC. The chromatograms shown in Fig. 4 indicated that THF and some ILs e.g. [C1 mim][MSO4 ] had a different extraction selectivity. It is apparent that MAE using THF led to an efficient extraction yield of norstictic acid (11.3%) but along with other substances in comparison to a more selective IL extraction (Fig. 4). The extraction yield of norstictic acid reached 3.2% with [C1 mim][MSO4 ] using the IL-MAE approach. Compared to HE using THF, the proposed process was 3fold less efficient (3.2–9.0%) (Table 2) but was compensated by a drastic reduction of the total extraction time (from 2 h to 5 min). ILs were also used in HE during 2 h and [C2 mim][ESO4 ] appeared to be the most effective with a 3.8% yield. As shown in Fig. 4, the purity of norstictic acid recovered from these two IL extracts is similar but in favor of [C1 mim][MSO4 ] in regard to the extraction time. Except for [C2 mim][ESO4 ], IL-MAE process was 1.5- to 3-fold more efficient than IL-HE (Fig. 3). These results suggested that most of ILs were more efficient under microwaves, due to their ability to transfer microwave energy in a good manner. Nevertheless MAE using THF was around 4-fold more efficient (11.3% yield and 97.8% ratio of norstictic acid in total extract) than IL-MAE (Table 2). This difference could be explained by a partial recovery of norstictic acid from ILs. Therefore, the recovery process of norstictic acid from ILs has to be optimized along with further extraction conditions. Despite this partial recovery, the best advantages of ILs in comparison to THF are their selective extraction, possible regeneration and no flammability. These ILs are also more safe than THF

Fig. 4. HPLC chromatograms of extracts using tetrahydrofuran under MAE (A), raw extract of [C1 mim][MSO4 ] under MAE (B), precipitates of [C1 mim][MSO4 ] under MAE (C), and [C2 mim][ESO4 ] under HE (D), respectively. Peak: norstictic acid-NA; column: Spherisorb C18 column (150 mm × 4.60 mm, 5 ␮m); mobile phase A (methanol) and B (water–acetic acid 0.6%): 60/40; flow rate: 0.8 mL min−1 ; UV wavelength: 239 nm.

74

S. Bonny et al. / Analytica Chimica Acta 707 (2011) 69–75

Table 2 Comparative study of extraction efficiency of norstictic acid using different extraction methods. Proposed method

Conventional methods

IL-MAE [C1 mim][MSO4 ]

Heat-reflux extraction (THF)

Regular MAE (THF)

Yield (mean ± S.D %)

Ratio in total yielda (mean ± S.D %)

Yield (mean ± S.D %)

Ratio in total yielda (mean ± S.D %)

Yield (mean ± S.D %)

Ratio in total yielda (mean ± S.D %)

3.2 ± 0.5

28.0 ± 0.3

9.0 ± 0.3

78.5 ± 6.2

11.3 ±1

98.7 ± 8.4

a

This ratio is the ratio of norstictic acid in total extract and was determined using the mean of total mass of norstictic acid in P. pseudocorallina which was 114.4 ± 5.6 mg g−1 .

Fig. 5. TLC profiles with UV detection of lichen metabolites in the [C1 mim][MSO4 ] extracts. (A) variolaric acid (VA), ␣-alectoronic acid (AA) and lecanoric acid (LA) from Ochrolechia parella (B) picrolichenic acid (PA) and derivatives from Pertusaria amara. Mobile phase: toluene–ethyl acetate–formic acid 70/25/5. UV detection at 270 nm.

because they do not form highly-explosive peroxides on storage in air.

3.4. Recovery of norstictic acid and regeneration of ionic liquids The general extraction process of norstictic acid and the potential regeneration of ionic liquid solvents are depicted in Fig. 2. Previous studies proved that some organic molecules could be extracted from ionic liquids with organic solvent [30]. We thus used water as co-extractant for solubilizing partially the extract and various organic solvents as extractants, the co-extractant leading the passage of products towards the extractant. This procedure was studied for [C1 mim][MSO4 ], [C2 mim][ESO4 ] and [C3 mimOH][NTf2 ] using diethyl ether, acetone and ethyl acetate to find the best extractant. This one must not be miscible with water and rather polar to extract most of the compounds. We chose the diethyl ether instead of the ethyl acetate extractant because of the lowest amount of IL remaining in dry extracts. The following addition of acetone/water mixture allowed us to elaborate a process to recover the compounds of interest by precipitation and to regenerate ILs. The extraction yields were then evaluated from precipitates using HPTLC coupled to a spectrophotodensitometer. Concerning the more hydrophobic ILs such as [C4 mim][NTf2 ] the recovery of NA was difficult as this IL is not soluble in water. We demonstrated here that the miscibility of ILs with water had a positive influence on their regeneration and the recovery of analytes. The possibility to regenerate more hydrophilic ILs by easy evaporation of water/acetone fraction could be in favor of this process in comparison to classical organic solvents.

[C1 mim][MSO4 ] extracted the major compounds of P. amara (picrolichenic acid and derivatives), which belong to the group of depsones, as well as it extracted variolaric acid, ␣-alectoronic acid (depsidones) and lecanoric acid (depside) from O. parella (Fig. 5). It was found that [C1 mim][MSO4 ] was an IL able to extract a great variety of lichen compounds such as depsides, depsidones and despones without degradation. Compared to HRE using THF, the IL based process thus appears to be a soft extractive method, what is particularly advantageous for the study of the depsides which are common in lichens but also among the most easily hydrolyzed molecules.

4. Conclusion In this work, ILs were proved to be alternative solvents for the extraction of major lichen compounds even if the recovery from ILs is not complete. They also allowed extraction to be carried on with a small quantity of raw material (0.25 g dried lichen). The cations and anions due to their difference in their physiochemical properties such as ability to interact with ␲-systems of the analytes and water miscibility had influences on the extraction efficiency. The best results were thus obtained using sulfate anions ([C1 mim][MSO4 ] and [C21 mim][ESO4 ]) associated to the imidazolium cation bearing a short chain related to a low hydrophobicity. Compared with conventional HE extraction procedures, the proposed approach provides sufficient extraction yields in a much shorter extraction time. Further experiments are in progress to improve the total recovery of compounds from IL extracts and to determine the influence of the sulfate anions on the extraction efficiency.

3.5. Extension of the method to other crustose lichens Acknowledgements In the light of previous results, we investigated the IL-MAE extraction process with other crustose lichens (P. amara and O. parella) containing different secondary metabolites [17]. Assays were performed under microwave irradiations with the most efficient IL which was [C1 mim][MSO4 ] (Fig. 5).

We thank the team of Prof. J.P. Bazureau (Rennes, France) for the use of the microwave apparatus and for their helpful comments. We are also deeply grateful to Isabelle Rouaud and Aurelie Sauvager for their helpful assistance.

S. Bonny et al. / Analytica Chimica Acta 707 (2011) 69–75

References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15]

J.S. Wilkes, Green Chem. 4 (2002) 73. A. Berthod, M.J. Ruiz-Angel, S. Carda-Broch, J. Chromatogr. A 1184 (2008) 6. J. Liu, J.A. Jonsson, G. Jiang, Anal. Chem. 24 (2005) 20. S.V. Smirnova, I.I. Torocheshnikova, A.A. Formanovsky, I.V. Pletnev, Anal. Bioanal. Chem. 378 (2004) 1369. K.J. Kulacki, G.A. Lamberti, Green Chem. 10 (2008) 104. S. Studzinska, B. Buszewski, Anal. Bioanal. Chem. 393 (2009) 983. F.Y. Du, X.H. Xiao, X.J. Luo, G.K. Li, Talanta 78 (2009) 1177. C.-H. Ma, T.-T. Liu, L. Yang, Y.-G. Zu, S.-H. Wang, R.-R. Zhang, Anal. Chim. Acta 689 (2011) 110. W. Ma, Y. Lu, R. Hu, J. Chen, Z. Zhang, Y. Pan, Talanta 80 (2010) 1292. C. Yansheng, Z. Zhida, L. Changping, L. Qingshan, Y. Peifanga, U. Welz-Biermann, Green Chem. 13 (2011) 666. C. Van Haluwyn, M. Lerond, Guide des lichens, Paris, 1993. K. Müller, Appl. Microbiol. Biotech. 56 (2001) 9. J. Boustie, M. Grube, Plant Gen. Res. 3 (2005) 273. D. O‘Donovan, G. Roberts, M. Keogh, Phytochemistry 19 (1980) 2497. F. Lohezic-Le Devehat, S. Tomasi, J.A. Elix, A. Bernard, I. Rouaud, P. Uriac, J. Boustie, J. Nat. Prod. 70 (2007) 1218.

75

[16] S. Bonny, E. Hitti, J. Boustie, A. Bernard, S. Tomasi, J. Chromatogr. A 1216 (2009) 7651. [17] S. Huneck, I. Yoshimura, Identification of Lichen Substances, Berlin-Heidelberg, 1996. [18] M. Millot, S. Tomasi, K. Articus, I. Rouaud, A. Bernard, J. Boustie, J. Nat. Prod. 70 (2007) 316. [19] J.D. Holbrey, W.M. Reichert, R.P. Swatloski, G.A. Broker, W.R.S.K.R. Pitner, R.D. Rogers, Green Chem. 4 (2002) 407. [20] J. Fraga-Dubreuil, J.P. Bazureau, Tetrahedron Lett. 41 (2000) 7351. [21] J. Fraga-Dubreuil, Ph.D. Dissertation, University of Rennes 1, 2002. [22] S. Bonny, Ph.D. Dissertation, University of Rennes 1, 2010. [23] Z. Guo, B.-M. Lue, K. Thomasen, A.S. Meyerb, X. Xu, Green Chem. 9 (2007) 1362. [24] J.L. Anderson, J. Ding, T. Welton, D.W. Armstrong, J. Am. Chem. Soc. 124 (2002) 14247. [25] S. Chun, S.V. Dzyuba, R.A. Bartsch, Anal. Chem. 73 (2001) 3737. [26] J. Wang, Y. Pei, Y. Zhao, Z. Hu, Green Chem. 7 (2005) 196. [27] J.G. Huddleston, A.E. Visser, W.M. Reichert, H.D. Willauer, G.A. Broker, R.D. Rogers, Green Chem. (2001) 156. [28] X. Cao, X. Ye, Y. Lu, Y. Yu, W. Mo, Anal. Chim. Acta 640 (2009) 47. [29] Y. Asahina, H.T. Akagi, T.B. Wake, Ber. Dtsch. Chem. Ges. 68B (1935) 1130. [30] X. Han, D.W. Armstrong, Acc. Chem. Res. 40 (2007) 1079.