Determination & Isolation Of Organochlorine Pesticide Residues In Chinese Herbal Medicine Using Supercritical Fluid Extraction

Organochlorine pesticides (OCPs) were once assumed necessary to protect crops against the pests and to allow a spectacular increase in production as well as the eradication of illness such as malaria and typhus. As fat-soluble compounds, OCPs are cumulative, persistent and difficult to be metabolized in both biota and environment matrixes and eventually become hazardous to human health through the food chain. For all of these reasons, most OCPs have been banned in most of the developed countries since the 1970s and 1980s. However, some developing countries are still using them for agricultural and public purposes because of its low cost and versatility in controlling insects. The cultivation of ginseng usually takes years and requires the use of pesticides, such as benzene hexachloride (BHC), DDT and its metabolites to reduce pest damage, which possess long half-life, chronic adverse effect and bioaccumulation in biotic fat. Therefore, the risk to health due to the incidental intake of contaminants is highly possible. Supercritical fluid extraction has received increasing attention as the solution for these problems due to the distinct properties such as lower viscosity, higher diffusivity, adjustable density, in which CO₂ is the commonly used extraction solvent due to its low critical temperature and its non-toxic and non-flammable properties. A method involving four-factor-three-level orthogonal array design including extracting solvent component, particle size, solvent overflow recycle and time needed for the optimization of extracting nine organochlorine pesticides from ginseng root was developed using Soxhlet extraction followed by capillary gas chromatography–electron capture detection and mass spectrometric confirmation. The optimal extraction conditions were: particle size of 60-80 mesh; n-hexane/acetone volume ratio of 3: 1; overflow recycle of 7 min; extraction time of 12 h, at which the total OCPs content was 4.6530 mg/kg for ginseng sample with particle size of 20-40 mesh. A method involving the simultaneous extraction and clean-up of nine organochlorine pesticides from the actual ginseng was developed using supercritical fluid extraction (SFE) and off-line solid phase extraction cartridge. The target analytes extracted were collected with an off-line commercial solid phase extraction cartridge, recovered with eluting solvent and quantified by gas chromatography-electron capture detection together with mass spectrometric confirmation. The optimal extraction conditions was obtained from a series of experiments in which samples preloaded by 10-wt% ethanol-H2O solutions, 30 MPa extraction pressure, 60℃ extraction temperature, with C18 as trap adsorbent and n-hexane as eluting solvent. Although maximum recovery was obtained at the optimal condition, the OCPs mass balance indicated that residue in ginseng was still above the maxmium residue level (MRL) of 0.1 mg/kg. According to the results above, a new process was developed, in which the organochlorine pesticide residue was partly isolated via supercritical CO2 from ginseng root hair without ginsenosides loss at 333 K and 30 MPa with certain solvent as modifier firstly; and then, ginsenosides were extracted through the current commercial industrial process. Ginsenosides were quantified by high-performance liquid chromatography equipped with UV detector. Results approved the feasibility to decrease the organochlorine pesticides content to 0.1 mg/kg in the final product without any ginsenosides loss in this process. The model compounds used in this study isomers of benzene hexachloride includingα-BHC,β-BHC,δ-BHC and pentchloronitrobenzene (PCNB), which were detected in ginseng root, whose solubility were determined in supercritical carbon dioxide by using a continuous flow apparatus at 308.15,318.15 and 333.15K and for pressures from 100 to 30 MPa after proving the accuracy of the apparatus by measuring naphthalene solubility in supercritical CO2. The experimental data were correlated with four empirical equations and gave better agreement. In the end, solubilities of the studied OCPs in CO2 and CO2 density were predicted by Peng-Ronbison EOS. To study the SFE kinetics, a neural network was developed to predict yields of the course of supercritical fluid extraction of organochlorine pesticides from ginseng. It consisted in a feed-forward multilayer neural network, conveniently trained with the back-propagation algorithm. Experiment data were divided into training samples and validation samples, inputs to the neural net were: pressure, temperature, extraction time and CO2 flow rate. The system provided a single output (the yields extracted) as the only system response. The network architecture was kept to minimum complexity by carefully choosing the number of units of the hidden layer. Predictions made by the selected artificial neuron net architecture successfully agreed with the experimental data. Additionally, based on some reasonable assumptions, a kinetical model was established and simulated by SIMULINK softpackage.