Electromembrane extraction (EME) is a principle for sample preparation, evolved from the principle of liquid-liquid extraction. In EME an electrical field facilitates the extraction of target analytes from an aqueous sample, through a supported liquid membrane (SLM), and into an acceptor solution . The SLM typically comprises approximately 10 µL of organic solvent (immiscible with water) immobilized in the pores of a polypropylene membrane. EME of basic analytes, as protonated species, occurs from neutral or acidified samples and using neutral or acidified aqueous buffer as acceptor solution. In such cases, the negative electrode is located in the acceptor solution and the SLM is typically 2-nitrophenyl octyl ether. EME of acidic analytes occur under neutral or alkaline conditions in the sample and acceptor solution, and with the positive electrode located in the acceptor solution (reversed polarity). EME is a highly selective method, due to the chemical composition of the SLM, and due to direction and magnitude of the electrical field. In addition, EME can provide pre-concentration of target analytes. EME is under commercial development, and the equipment is very close to the market. In the current work, we have tested a new prototype device based on EME. This device comprises electrically conducting vials for housing sample and acceptor solutions. Thus, unlike traditional EME, with platinum electrodes inserted in the sample and acceptor solution, the prototype relies on coupling the electrical field through the conducting containers. In this thesis, we demonstrate optimization for pethidine, haloperidol, nortriptyline, methadone, and loperamide with the prototype device. These substances are basic drugs selected as model compounds. The following factors had an impact on recovery, and were optimized in this work: extraction time, volume of sample and acceptor solution, voltage, shaking and volume of SLM. The model analytes were extracted with the use of the prototype device. Separation, detection, and quantification of analytes from acceptor solution were performed using HPLC-UV. Using a “Design of experiment” approach, optimization was carried out through several steps including factor screening with fractional factorial design, the path of steepest ascent, and response surface methodology. The last step included testing the robustness of the statistical model. The prototype set-up provided a very stable extraction system. It responded to changes in operational parameters according to theory while providing excellent repeatability (RSD < 15%). Average recovery of 88% for five analyte molecules was achieved under the following conditions: voltage - 158 V, shaking rate 1500 rpm, volumes of donor and acceptor 905 µl, volume of SLM 13 µl and extraction time of 24 minutes.