Benzodiazepines (BZD) and Z-hypnotics are frequently analyzed in forensic laboratories, and in 2012, the designer benzodiazepines (DBZD) emerged on the illegal drug scene. DBZD represent a particular challenge demanding new analytical methods. In this work, parallel artificial liquid membrane extraction (PALME) is used for sample preparation of DBZD, BZD, and Z-hypnotics in whole blood prior to UHPLC-MS/MS analysis. PALME of BZD, DBZD, and Z-hypnotics was performed from whole blood samples, and the analytes were extracted across a supported liquid membrane (SLM) and into an acceptor solution of dimethyl sulfoxide and 200 mM formic acid (75:25, v/v). The method was validated according to EMA guidelines. The method was linear throughout the calibration range (R > 0.99). Intra- and inter-day accuracy and precision, as well as matrix effects, were within the guideline limit of ± 15%. LOD and LLOQ ranged from 0.10 to 5.0 ng mL and 3.2 to 160 ng mL, respectively. Extraction recoveries were reproducible and above 52%. The method was specific, and the analytes were stable in the PALME extracts for 4 and 10 days at 10 and - 20 °C. No carry-over was observed within the calibration range. PALME and UHPLC-MS/MS for the determination of DBZD, BZD, and Z-hypnotics in whole blood are a green and low-cost alternative that provides high sample throughput (96-well format), extensive sample clean-up, good sensitivity, and high reproducibility. The presented method is also the first method incorporating analysis of DBZD, BZD, and Z-hypnotics in whole blood in one efficient analysis. Graphical abstract.

Complexation-mediated electromembrane extraction (EME) of highly polar basic drugs (log P < -1) was investigated for the first time with the catecholamines epinephrine, norepinephrine, and dopamine as model analytes. The model analytes were extracted as cationic species from urine samples (pH 4), through a supported liquid membrane (SLM) comprising 25 mM 4-(trifluoromethyl)phenylboronic acid (TFPBA) in bis(2-ethylhexyl) phosphite (DEHPi), and into 20 mM formic acid as acceptor solution. EME was performed for 15 min, and 50 V was used as extraction voltage across the SLM. TFPBA served as complexation reagent, and selectively formed boronate esters by reversible covalent binding with the model analytes at the sample/SLM interface. This enhanced the mass transfer of the highly polar model analytes across the SLM, and EME of basic drugs with log P in the range -1 to -2 was shown for the first time. Meanwhile, most matrix components in urine were unable to pass the SLM. Thus, the proposed concept provided highly efficient sample clean-up and the system current across the SLM was kept below 50 μA. Finally, the complexation-mediated EME concept was combined with ultra-high performance liquid chromatography coupled to tandem mass spectrometry and evaluated for quantification of epinephrine and dopamine. Standard addition calibration was applied to a pooled human urine sample. Calibration curves using standards between 25 and 125 μg L gave a high level of linearity with a correlation coefficient of 0.990 for epinephrine and 0.996 for dopamine (N = 5). The limit of detection, calculated as three times signal-to-noise ratio, was 5.0 μg L for epinephrine and 1.8 μg L for dopamine. The repeatability of the method, expressed as coefficient of variation, was 13% (n = 5). The proposed method was finally applied for the analysis of spiked pooled human urine sample, obtaining relative recoveries of 91 and 117% for epinephrine and dopamine, respectively.

Parallel artificial liquid membrane extraction (PALME) was combined with ultra-high performance liquid chromatography-mass spectrometry (UHPLC-MS) and the potential for screening of new psychoactive substances (NPS) was investigated for the first time. PALME was performed in 96-well format comprising a donor plate, a supported liquid membrane (SLM), and an acceptor plate. Uncharged NPS were extracted from plasma or whole blood, across an organic SLM, and into an aqueous acceptor solution, facilitated by a pH gradient. MDAI (5,6-methylenedioxy-2-aminoindane), methylone, PFA (para-fluoroamphetamine), mCPP (meta-chlorophenylpiperazine), pentedrone, methoxetamine, MDPV (methylenedioxypyrovalerone), ethylphenidate, 2C-E (2,5-dimethoxy-4-ethylphenethylamine), bromo-dragonfly, and AH-7921 (3,4-dichloro-N-{[1-(dimethylamino)cyclohexyl]methyl}benzamide) were selected as representative NPS. Optimization of operational parameters was necessary as the NPS were novel to PALME, and because PALME was performed from whole blood for the very first time. In the PALME method developed for plasma, NPS were extracted from a 250μL alkalized donor solution consisting of 125μL plasma sample, 115μL 40mM NaOH, and 10μL internal standard. In the PALME method from whole blood, the 250μL alkalized donor solution consisted of 100μL whole blood, 50μL deionized water, 75μL 80mM NaOH, and 25μL internal standard. In both methods, extraction was accomplished across an SLM of 5μL dodecyl acetate with 1% trioctylamine (w/w), and further into an acidic acceptor solution of 50μL 20mM formic acid. The extraction was promoted by agitation at 900rpm and was carried out for 120min. Method validation was performed and the following parameters were considered: linearity, limits of quantification (LOQ), intra- and inter-day precision, accuracy, extraction recoveries, carry-over, and matrix effects. The validation results were in accordance with FDA guidelines.

Ultra-HPLC-MS/MS analysis of the donor solutions revealed that the phospholipids principally remained in the plasma samples. This proved that the phospholipids did not migrate in the electrical field and they were prevented from penetrating the supported liquid membrane.

Recoveries more than 80% were obtained for analytes with pKa > 2, whereas EME was less efficient for substances with pKa < 2. The latter was trapped in the supported liquid membrane due to less acidic pH conditions in the acceptor solution close to the supported liquid membrane. EME followed by UHPLC-MS/MS analysis was evaluated from human plasma, and the results were in compliance with EMA guidelines. Both electrokinetic migration and passive diffusion contributed to mass transfer when performing EME of weakly basic analytes.