Microfluidic lab-on-chip advances for liquid–liquid extraction process studies

The ever-increasing complexity of industrial product recycling calls for more efficient purification processes such as liquid–liquid (L/L) extraction. Because of the high complexity of L/L extraction, optimizing a large-scale extraction is both time and resource consuming and can only be justified to solve high volume and value purification problems. It is therefore difficult to apply to small scale and highly variable waste material influx. We believe using a fully automated and integrated microfluidic approach will enable fast and cost-effective studies of L/L extraction processes. This review presents an overview of L/L metal extraction performed using microfluidics platforms. We first give an overview of the extraction methods. We then review the most relevant characterization methods integrated with such platforms.     

Lipophilic ternary complexes in liquid–liquid extraction of trivalent lanthanides

The formation of ternary complexes between lanthanide ions [Nd(III) or Eu(III)], octyl(phenyl)-N,N-diisobutyl-carbamoylmethylphosphine oxide (CMPO), and bis-(2-ethylhexyl)phosphoric acid (HDEHP) was probed by liquid–liquid extraction and spectroscopic techniques. Equilibrium modeling of data for the extraction of Nd(III) or Eu(III) from lactic acid media into n-dodecane solutions of CMPO and HDEHP indicates the predominant extracted species are of the type [Ln(AHA)₂(A)] and [Ln(CMPO)(AHA)₂(A)], where Ln?=?Nd or Eu and A represents the DEHP^? anion. FTIR (for both Eu and Nd) and visible spectrophotometry (in the case of Nd) indicate the formation of the [Ln(CMPO)(A)₃] complexes when CMPO is added to n-dodecane solutions of the LnA₃ compounds. Both techniques indicate a stronger propensity of CMPO to complex Nd(III) versus Eu(III).     

Alkylsulfate-based ionic liquids in the liquid–liquid extraction of aromatic hydrocarbons

The (liquid + liquid) equilibrium data (LLE) for the extraction of toluene from heptane with different ionic liquids (ILs) based on the alkylsulfate anion (R-SO₄) was determined at T = 313.2 K and atmospheric pressure. The effect of more complex R-SO₄ anions on capacity of extraction and selectivity in the liquid–liquid extraction of toluene from heptane was studied. The ternary systems were formed by {heptane + toluene + 1,3-dimethylimidazolium methylsulfate ([mmim][CH₃SO₄]), 1-ethyl-3-methylimidazolium hydrogensulfate ([emim][HSO₄]), 1-ethyl-3-methylimidazolium methylsulfate ([emim][CH₃SO₄]), or 1-ethyl-3-methylimidazolium ethylsulfate ([emim][C₂H₅SO₄])}. The degree of quality of the experimental LLE data was ascertained by applying the Othmer–Tobias correlation. The phase diagrams for the ternary systems were plotted, and the tie lines correlated with the NRTL model compare satisfactorily with the experimental data.     

A simple novel configuration for in-vial microporous membrane liquid–liquid extraction

A novel arrangement for microporous membrane liquid–liquid extraction from the aqueous donor phase to the organic acceptor phase within a micro-vial, which is compatible with the chromatograph autosampler is presented. The device consisted of a stoppered glass micro-vial containing the organic solvent where the septum of the screw stopper was replaced by a sized piece of membrane which is hermetically assembled to the volumetric flask containing the aqueous donor solution. The placement of the membrane in alternative contact with the solutions was achieved by orbital agitation. As a preliminary study, 2-ethylhexyl 4-(dimethylamino)benzoate has been determined (limit of quantification 0.11 μg L⁻¹, precision 7.4%). The small quantity of organic solvent used, the achieved sample cleanup, and the minimal handling and risk of cross-contamination are significant operational advantages.     

Analysis of PAH compounds in soil with on-line coupled pressurised hot water extraction–microporous membrane liquid–liquid extraction–gas chromatography

Pressurised hot water extraction (PHWE) was coupled on-line with microporous membrane liquid-liquid extraction (MMLLE) and gas chromatography (GC) in the analysis of polycyclic aromatic hydrocarbon (PAH) compounds in soil. The MMLLE serves as a trapping device after the PHWE. Water from PHWE is directed to the donor side of the membrane unit and the analytes are extracted to the acceptor solution on the other side of the membrane. The role of MMLLE is to clean and concentrate the extract, which is then transferred on-line to the GC via a sample loop and an on-column interface using partially concurrent solvent evaporation. Separate optimisation of MMLLE and simulations of the PHWE-MMLLE connection were carried out before the actual on-line coupling. After optimisation of the whole on-line system, the efficiencies of the PHWE-MMLLE-GC and PHWE-solid-phase trap extractions were compared. The PHWE-MMLLE-GC method allowed on-line analysis of soil samples. The method was linear, with limits of detection in the range 0.05-0.13 ng and limits of quantification 0.65-1.66 microg g(-1). Comparison of the results with those obtained by other techniques confirmed the good performance.     

Ionic liquid based dispersive liquid–liquid microextraction for the extraction of pesticides from bananas

This paper describes a dispersive liquid–liquid microextraction (DLLME) procedure using room temperature ionic liquids (RTILs) coupled to high-performance liquid chromatography with diode array detection capable of quantifying trace amounts of eight pesticides (i.e. thiophanate-methyl, carbofuran, carbaryl, tebuconazole, iprodione, oxyfluorfen, hexythiazox and fenazaquin) in bananas. Fruit samples were first homogenized and extracted (1 g) with acetonitrile and after suitable evaporation and reconstitution of the extract in 10 mL of water, a DLLME procedure using 1-hexyl-3-methylimidazolium hexafluorophosphate ([C₆MIM][PF₆]) as extraction solvent was used. Experimental conditions affecting the DLLME procedure (sample pH, sodium chloride percentage, ionic liquid amount and volume of disperser solvent) were optimized by means of an experimental design. In order to determine the presence of a matrix effect, calibration curves for standards and fortified banana extracts (matrix matched calibration) were studied. Mean recovery values of the extraction of the pesticides from banana samples were in the range of 69–97% (except for thiophanate-methyl and carbofuran, which were 53–63%) with a relative standard deviation lower than 8.7% in all cases. Limits of detection achieved (0.320–4.66 μg/kg) were below the harmonized maximum residue limits established by the European Union (EU). The proposed method, was also applied to the analysis of this group of pesticides in nine banana samples taken from the local markets of the Canary Islands (Spain). To the best of our knowledge, this is the first application of RTILs as extraction solvents for DLLME of pesticides from samples different than water.     

Pesticide extraction from table grapes and plums using ionic liquid based dispersive liquid–liquid microextraction

Room temperature ionic liquids (RTILs) have been used as extraction solvents in dispersive liquid–liquid microextraction (DLLME) for the determination of eight multi-class pesticides (i.e. thiophanate-methyl, carbofuran, carbaryl, tebuconazole, iprodione, oxyfluorfen, hexythiazox, and fenazaquin) in table grapes and plums. The developed method involves the combination of DLLME and high-performance liquid chromatography with diode array detection. Samples were first homogenized and extracted with acetonitrile. After evaporation and reconstitution of the extract in water containing sodium chloride, a quick DLLME procedure that used the ionic liquid 1-hexyl-3-methylimidazolium hexafluorophosphate ([C₆MIM][PF₆]) and methanol was developed. The RTIL dissolved in a very small volume of acetonitrile was directed injected in the chromatographic system. The comparison between the calibration curves obtained from standards and from spiked sample extracts (matrix-matched calibration) showed the existence of a strong matrix effect for most of the analyzed pesticides. A recovery study was also developed with five consecutive extractions of the two types of fruits spiked at three concentration levels. Mean recovery values were in the range of 72–100% for table grapes and 66–105% for plum samples (except for thiophanate-methyl and carbofuran, which were 64–75% and 58–66%, respectively). Limits of detection (LODs) were in the range 0.651–5.44 μg/kg for table grapes and 0.902–6.33 μg/kg for plums, representing LODs below the maximum residue limits (MRLs) established by the European Union in these fruits. The potential of the method was demonstrated by analyzing 12 commercial fruit samples (six of each type).     

Non-porous membrane-assisted liquid–liquid extraction of UV filter compounds from water samples

A method for the determination of nine UV filter compounds [benzophenone-3 (BP-3), isoamyl methoxycinnamate, 4-methylbenzylidene camphor, octocrylene (OC), butyl methoxydibenzoylmethane, ethylhexyl dimethyl p-aminobenzoate (OD-PABA), ethylhexyl methoxycinnamate (EHMC), ethylhexyl salicylate and homosalate] in water samples was developed and evaluated. The procedure includes non-porous membrane-assisted liquid–liquid extraction (MALLE) and LC–atmospheric pressure photoionisation (APPI)–MS/MS. Membrane bags made of different polymeric materials were examined to enable a fast and simple extraction of the target analytes. Among the polymeric materials tested, low- and high-density polyethylene membranes proved to be well suited to adsorb the analytes from water samples. Finally, 2 cm length tailor-made membrane bags were prepared from low-density polyethylene in order to accommodate 100 μL of propanol. The fully optimised protocol provides recoveries from 76% to 101% and limits of detection (LOD) between 0.4 ng L⁻¹ (OD-PABA) and 16 ng L⁻¹ (EHMC). The interday repeatability of the whole protocol was below 18%. The effective separation of matrix molecules was proved by only marginal matrix influence during the APPI-MS analysis since no ion suppression effects were observed. During the extraction step, the influence of the matrix was only significant when non-treated wastewater was analysed. The analysis of lake water indicated the presence of seven UV filter compounds included in this study at concentrations between 40 ng L⁻¹ (BP-3) and 4381 ng L⁻¹ (OC). In non-treated wastewater several UV filters were also detected at concentration levels as high as 5322 ng L⁻¹ (OC).     

High-performance liquid chromatographic determination of diltiazem in human plasma after solid-phase and liquid–liquid extraction

A selective, sensitive, and accurate high-performance liquid chromatographic method for determination of diltiazem in plasma samples has been developed and validated. The effects of mobile phase composition, buffer concentration, mobile phase pH, and concentration of organic modifiers on retention of diltiazem and internal standard were investigated. Solid-phase and liquid–liquid extraction were examined and proposed for isolation of the drug and elimination of endogenous plasma interferences. A 5 m Lichrocart Lichrospher 60 RP-select B chromatographic column was used; the mobile phase was acetonitrile–0.025 mol L^–1 KH₂PO₄ (pH 5.5), 35:65 ( v / v) at a flow-rate of 1.5 mL min^–1. The detection wavelength was 215 nm. The calibration plots were linear in the concentration range 20.0–500.0 ng mL^–1. The method has been implemented to monitor diltiazem levels in patient samples.     

Solvent extraction and extraction–voltammetric determination of phenols using room temperature ionic liquid

The phenolic compounds phenol, 4-nitrophenol, 2,4-dinitrophenol, 2,6-dinitrophenol, 1-naphthol, 2-naphthol, and 4-chlorophenol are extracted nearly quantitatively from aqueous solution into the room temperature ionic liquid 1-butyl-3-methylimidazolium hexafluorophosphate (BMImPF₆) in molecular form at pH<pK_a. Picric acid is extracted efficiently in anionic form. Recovery of pyrocatechol and resorcinol is much lower. The effect of pH, phenol concentration, and volume ratio of aqueous and organic phases were studied. Ionic liquid BMImPF₆ is shown to be suitable for extraction–voltammetric determination of phenols without back-extraction or addition of support electrolyte. The electrochemical window of BMImPF₆ at various electrodes was determined, and voltammetric oxidation of phenols and reduction of nitrophenols in BMImPF₆ was studied.     

Sensitive determination of nitrophenol isomers by reverse-phase high-performance liquid chromatography in conjunction with liquid–liquid extraction

A method for the highly sensitive determination of 2-, 3- and 4-nitrophenols was developed using reverse-phase high-performance liquid chromatography (RP-HPLC) with a UV photodiode array detector. Using a reverse-phase column and 40% aqueous acetonitrile as an eluent (i.e. isocratic elution), the integrated peak area of detector output was linear up to 300 mg/L and the detection limit was 150 µg/L. The sensitivity of this detection method was improved by pretreating the sample solutions with a solvent extraction procedure that makes use of the high partition coefficient of ethyl acetate (EA)/water system. To find an optimum condition for the extraction procedure, this process was simulated by plotting the concentration of nitrophenol extracted in organic solvent against the volume multiplication factor at various partition coefficient of solute. This simulation demonstrated that EA is a superior extractant to other organic solvents. With the newly developed method, the detection limit was extended to 0.3 µg/L. This method offers fast, reliable and more sensitive determination of nitrophenol isomers than any other HPLC method.     

Preconcentration technique for nonylphenol using cellulose cotton with homogenous liquid–liquid extraction for liquid chromatographic analysis

A powerful preconcentration method for nonylphenol (NP) has been developed for liquid-chromatography by combining the use of cellulose cotton (solid-phase extraction) with homogeneous liquid–liquid extraction. A 100 ml of sample solution was preconcentrated using cotton, and the eluate obtained (acetonitrile; 5 ml) was further preconcentrated to 50 l within 10 min using a homogeneous liquid–liquid extraction method (volume ratio, 2,000-fold; 100 ml 50 l). The sample concentration increases from preconcentration was 1,599-fold, and NP was extracted into the sedimented phase at 80%. The proposed method was applied to high performance liquid chromatography with fluorescence detection (FL/HPLC); the lowest determination limit obtained was 1.0×10^–9 mol l^–1.     

Liquid–liquid extraction of scandium with nalidixic acid in dichloromethane

The extraction behavior of nalidixic acid (HNA) in CH₂Cl₂ has been studied for various di- and trivalent metal ions such as Cu(II), Fe(II), Ni(II), Mn(II), Sb(II), Co(II), Sc(III), Y(III), Nd(III) and Eu(III) from aqueous buffer solutions of pH 1–7 with 0.1 mol dm⁻³ nalidixic acid in dichloromethane. Separation factors of Sc(III) from these metals has shown that its clean separation is possible at pH 3.4–4. The parameters affecting the extraction of Sc(III) were optimized. The composition of the extracted adduct was determined by slope analysis method that came out to be Sc(NA)₃. Extraction of Sc(III) was studied in the presence of various cations and anions. Among the anions studied only fluoride, citrate and oxalate have significant interference whereas, Fe(III) has reduced the extraction to 53% that can be removed by using ascorbic acid as reducing agent. The proposed extraction system proved good stability up to six extraction-stripping stages for the extraction of Sc(III).     

Determination of chlorpyrifos in soils using multi-throughput dynamic microwave-assisted extraction coupled online with salting-out-assisted liquid–liquid extraction followed by LC–MS/MS

A simple and rapid method based on multi-throughput dynamic microwave-assisted extraction coupled online with salting-out-assisted liquid–liquid extraction was developed for the analysis of chlorpyrifos in soil. First, the chlorpyrifos was extracted with acetonitrile aqueous (50%, v/v) under the action of microwave energy. Then the obtained extract was separated clearly and easily into an acetonitrile phase and an aqueous phase with the assistance of ammonium acetate. The acetonitrile phase containing chlorpyrifos was concentrated and determined by liquid chromatography–tandem mass spectrometry. The effects of parameters on extraction efficiency including microwave power, extraction solvent, volumes and flow rate of extraction solvent, sample pH, types and amount of salt were studied and optimised. To eliminate the matrix effect, validation of the method was carried out using the matrix-based calibration curve. The limits of detection and quantification for chlorpyrifos were 0.17 and 0.5 ng g^?1, respectively. The proposed method was applied to analyse chlorpyrifos in five soil samples and verified by the recovery test. The recoveries of chlorpyrifos at three spiked levels (5, 50, 200 ng g^?1) were in the range of 90.0–100.5%, with relative standard deviations varying from 1.3% to 5.7%. Compared with the methods reported previously, the proposed method can simplify the operation procedure and reduce solvent consumption in sample pretreatment.     

Determination of ochratoxin A in pig tissues by liquid–liquid extraction and clean-up and high-performance liquid chromatography

A fast, simple, and sensitive HPLC–FD method is described for determination of ochratoxin A (OTA) in pig kidney and muscle; a small mass (<2.5 g) of sample and a relatively small volume (<15 mL) of a non-halogenated extraction solvent are required. Ochratoxin B, systematically absent from all the samples investigated, was used as internal standard. Liquid–liquid partition was used for sample clean-up. Recoveries at the 1 ng g^–1 level were 86±15% and 74±8% for kidney and muscle, respectively, and detection limits were 0.14 and 0.15 ng g^–1. Clean-up by solid-phase extraction (SPE) is required for pig liver. A survey of the OTA content of tissues of pigs slaughtered in southern Italy revealed that 52 out of 54 analysed samples were contaminated; the OTA concentration in kidney ranged between 0.26 and 3.05 ng g^–1. The effect of measurement precision on compliance with legal limits is also discussed.     

Semi-automated salt-assisted liquid–liquid extraction coupled to high-performance liquid chromatography to determine three aromatic hydrocarbons in aqueous samples

In this study, a new device for semi-automated salt-assisted liquid–liquid extraction was designed and coupled with high-performance liquid chromatography (HPLC) to determine three aromatic hydrocarbons in aqueous samples. In order to evaluate the performance of the designed device, three aromatic hydrocarbons including 2-naphthol, naphthalene and anthracene were selected as model analytes. Sample solution, extraction solvent and salt solution using separate channels were transferred to a sample holder, respectively. These three components were mixed using a magnetic stirrer. After stirrer stopping, the aqueous and organic phases were separated and organic layer transferred to the injection loop of HPLC system. Optimization process was achieved using response surface methodology by Design-Expert software. A central composite design was used to optimize the main parameters including pH (A), stirrer time (B), organic solvent volume (C) and salt concentration (D). The limit of quantitation for 2-naphthol, naphthalene and anthracene was 15.0, 25.0 and 1.0 ng mL^?1, respectively. Under the optimum conditions, obtained recoveries for three analytes were in the range of 76.0–96.2% with relative standard deviation less than 8.2%. The salt-assisted liquid–liquid extraction method using the proposed device has been successfully used for the analysis of real samples containing studied analytes in various matrices.     

Determination of resveratrol in wine by photochemically induced second-derivative fluorescence coupled with liquid–liquid extraction

Basic studies on the photochemical behaviour of trans-resveratrol and its photoproduct are reported. Photometrically and fluorimetrically calculated acidity constants of the former were determined. The usefulness of the determination of resveratrol by photochemically induced fluorescence and second-derivative photochemically induced fluorescence was also examined. The very weakly fluorescent trans-resveratrol is converted into a highly fluorescent photoproduct by irradiating hydroethanolic solutions of trans-resveratrol containing 40% v/v of ethanol for 60 s with intense UV radiation. The photoproduct presents excitation and emission maxima centred at 260 nm, and 364 and 382 nm, respectively. Under these conditions, a linear relationship between fluorescence intensity and trans-resveratrol concentration was found between 6.6 and 66 ng mL⁻¹. Optimum conditions for the extraction of trans-resveratrol from an aqueous phase at pH 5.0 with diethylether were a phase ratio (aqueous/organic) of 2, a shaking time of 60 s and a buffer concentration of 0.15 mol L⁻¹. An extraction recovery of 100% was reached under these conditions. The optimized extraction procedure was applied to the analysis of resveratrol in wine samples, employing the amplitude between 356 and 364 nm of the second-derivative photoinduced emission spectrum as analytical signal. It was found that there is not matrix effect and recoveries around 100% were obtained at different fortification levels.     

Hydrophobic polymer monoliths as novel phase separators: Application in continuous liquid–liquid extraction systems

Hydrophobic macroporous polymer monoliths are shown to be interesting materials for the construction of “selective solvent gates”. With the appropriate surface chemistry and porous properties the monoliths can be made permeable only for apolar organic solvents and not for water. Different poly(butyl methacrylate-co-ethylene dimethacrylate) (BMA-EDMA) and poly(styrene-co-divinylbenzene) (PS-DVB) monoliths prepared with tailored chemistries and porosities were evaluated for this purpose. After extensive characterization, the PS-DVB monoliths were selected due to their higher hydrophobicity and their more suitable flow characteristics. BMA-EDMA monoliths are preferred for mid-polarity solvents such as ethyl acetate, for which they provide efficient separation from water. Breakthrough experiments confirmed that the pressures necessary to generate flow of organic solvents through PS-DVB monoliths were substantially lower than for water. A phase separator was constructed using the monoliths as the flow selector. This device was successfully coupled on-line with a chip-based continuous liquid–liquid-extraction (LLE) system with segmented flow. Efficient separation of different solvents was obtained across a wide range of flow rates (0.5–4.0 mL min⁻¹) and aqueous-to-organic flow ratios (β = 1–10). Good robustness and long life-time were also confirmed. The suitability of the device to perform simple, cheap, and reliable phase separation in a continuous LLE system prior to gas-chromatographic analysis was proven for some selected real-life applications.     

Development of deep eutectic solvent–based microwave-assisted extraction combined with temperature controlled ionic liquid–based liquid phase microextraction for extraction of aflatoxins from cheese samples

In this study, for the first time, a deep eutectic solvent-based microwave-assisted extraction was combined with ionic liquid–based temperature controlled liquid phase microextraction for the extraction of several aflatoxins from cheese samples. Briefly, the analytes are extracted from cheese sample (3 g) into a mixture of 1.5 mL choline chloride:ethylene glycol deep eutectic solvent and 3.5 mL deionized water by exposing to microwave irradiations for 60 s at 180 W. The liquid phase was taken and mixed with 55 μL 1-hexyl-3-methylimidazolium hexafluorophosphate. By cooling the solution in the refrigerator centrifuge, a turbid state was obtained and the analytes were extracted into the ionic liquid droplets. The analytes were determined by high-performance liquid chromatography equipped with fluorescence detector. Low limits of detection (9–23 ng kg^–1) and quantification (30–77 ng kg^–1), high extraction recovery (66%–83%), acceptable enrichment factor (40–50), and good precision (relative standard deviations ≤ 5.2%) were obtained using the offered approach. These results reveal the high extraction capability of the method for determination of aflatoxins in the cheese samples. In this method, there was no need for organic solvents and it can be considered as green extraction method.     

Determination of carbohydrates in tobacco by pressurized liquid extraction combined with a novel ultrasound-assisted dispersive liquid–liquid microextraction method

A novel derivatization-ultrasonic assisted-dispersive liquid–liquid microextraction (UA-DLLME) method for the simultaneous determination of 11 main carbohydrates in tobacco has been developed. The combined method involves pressurized liquid extraction (PLE), derivatization, and UA-DLLME, followed by the analysis of the main carbohydrates with a gas chromatography-flame ionization detector (GC-FID). First, the PLE conditions were optimized using a univariate approach. Then, the derivatization methods were properly compared and optimized. The aldononitrile acetate method combined with the O-methoxyoxime-trimethylsilyl method was used for derivatization. Finally, the critical variables affecting the UA-DLLME extraction efficiency were searched using fractional factorial design (FFD) and further optimized using Doehlert design (DD) of the response surface methodology. The optimum conditions were found to be 44 μL for CHCl₃, 2.3 mL for H₂O, 11% w/v for NaCl, 5 min for the extraction time and 5 min for the centrifugation time. Under the optimized experimental conditions, the detection limit of the method (LODs) and linear correlation coefficient were found to be in the range of 0.06–0.90 μg mL⁻¹ and 0.9987–0.9999. The proposed method was successfully employed to analyze three flue-cured tobacco cultivars, among which the main carbohydrate concentrations were found to be very different.