Extraction matrine from Radix Sopheorae Tonkinensis by non-supported liquid membrane extraction technology

2.1.1 Instruments

The results analysis was carried out using an Agilent 6890 N GC system coupled with the fused silica capillary column (Hp-5, 32 m, length × 0.32 mm i.d., 1.05 μm film thickness). Oven temperature followed the gradient: kept at the initial temperature of 40 °C for 1 min, and rose up to 220 °C at 40 °C/min, and then rose up again to 280 °C at 4 °C/min and held for 10 min. The interface line temperature was 220 °C. Helium was the carrier gas in constant flow fashion, the flow rate was 2.4 mL/min, and the column pressure was 35 kPa. Two microliters of each sample was injected in the GC system at a splitless mode.

For the MS analysis (VG70E-HF, Micromass Corporation), an EI source was used in the total scan mode between 30 and 600 m/z. The electron energy was 70 eV, the accelerated voltage was 6 kV, the ion source temperature was 200 °C, and the resolving power was 1000.

2.1.2 Regents

Radix Sophorae Tonkinensis were purchased from Tong Ren Tang Pharmacy (Baoding China); caffeine (self-made); matrine was obtained from Biotechnology Co., Ltd. (Shaanxi Zhongxin China); PVC sample tube (1.5 mL volume) and camphor, ethanol, trichloromethane, sodium hydroxide, phosphate and monosodium orthophosphate (NaH₂PO₄) were purchased from Chemical Agent Shop (Baoding, China). All the chemicals are of analytical grade. The water used in the experiments was doubly distilled water prepared in our lab.

2.2.1 Sample preparation

Twenty grams of Radix Sophorae Tonkinensis was wrapped with a filter paper and placed into the Soxhlet extractor. It was extracted by 150 mL ethanol–water solution (v:v, 80:20) to a colorless liquid, and ethanol was evaporated, left behind was the 28 mL pre-extracts for further extraction.

2.2.2 Standard solution preparation

One hundred milligram of matrine was weighed and dissolved in a 50 mL beaker and was then drained into a 100 mL volumetric flask and water was added to a total volume of 100 mL to make a 1000 mg L⁻¹ matrine solution. The solution was then diluted to the concentrations of 300, 250, 200, 150 and 100 mg L⁻¹, respectively. A 500 mg L⁻¹ caffeine solution was prepared the same way by weighing 25 mg caffeine and dissolved it by CHCL₃ in 50 mL volumetric flask, which was as the internal standard. 0.2 mL of the internal standard solution was mixed with 0.8 mL of matrine standard solutions to make the spiking solution. Two microliters of the spiked solution was injected into GC/MS and repeated three times. The calibration curve was calculated by using peak area ratios between matrine and caffeine. A linear relationship was obtained with a good correlation coefficient (Fig. 1 Y = 0.00596x − 0.00400, R = 0.991), when the matrine solution concentration was as axis X, and the ratio of matrine and caffeine chromatographic peak area was as axis Y.

Close

Figure 1

The calibration curve of matrine.

Export to PPT

2.2.3 Extraction procedure

PVC sample tubes (1.5 mL) were used as the test device. Phosphoric acid was added into PVC tubes and the density was adjusted between 1.55 and 1.80 by sodium dihydrogen phosphate. Then chloroform was added as a membrane solution forming the middle organic layer (density 1.49). At last, analyte solution (pH 13) was added at the top. After a while, the alkali solution in PVC tubes was removed and pooled together. All the samples were then extracted by chloroform, added 0.2 mL of the internal standard solution and then analyzed by the GC/MS. The acid solution in PVC tubes was also combined together, and pH was adjusted between 13 and 14 with solid sodium hydroxide. Then they were extracted by chloroform and added internal standard for GC/MS analysis.

3.1.1 Influence of extraction time

Experiments were carried out at room temperature, and using 300 mg L⁻¹ matrine aqueous solution as the donor phase. Fig. 2 shows that the amount of extracted matrine increased with the increased extraction time. In the first 8 h extraction, the rate of extraction was fast and close to 60%. Then it slowed down in the next 2–12 h. After 22 h, since all matrine in the donor phase had reached the acceptor phase, the amount of extracted matrine remained constant.

Close

Figure 2

The effects of extraction time on the extraction quantity.

Export to PPT

3.1.2 Influence of extraction temperature

Membrane extraction quantity is affected by immigration rate of extracts, which has a direct relationship with extraction temperature. Thus, extraction temperature is a direct influence factor of extraction efficiency. In the experiment, the extractions were carried out at six different temperatures (35, 40, 45, 50, 55, 60 °C) for 2 h using standard matrine solution (1000 mg L⁻¹) as the donor phase. Fig. 3 shows that the optimal extraction temperature is 45 °C, when the extraction amount achieved 0.65 mg. And the extraction amount decreased either below 45 °C or above it. The reason is that both diffusion coefficient and distribution coefficient could affect the extraction capacity. And the diffusion coefficient will increase with increased temperature, but the distribution coefficient will reduce (Ritcey and Ashbrook, 1984). Therefore, there is an optimal temperature to balance both of them perfectly and achieve a maximum amount of the extraction. Experiment was not carried out at a temperature more than 60 °C, because it exceeds the boiling point of chloroform which might give rise to an unstable extraction system.

Close

Figure 3

The effects of extraction temperature on the extraction quantity.

Export to PPT

3.1.3 Influence of the volume of the organic phase

Experiments were carried out with three different volumes of organic phase (0.2, 0.25, 0.3 mL) at 45 °C using standard matrine solution (1000 mg L⁻¹) as the donor phase. Sine experiments were conducted in PVC tubes where the bottom areas are the same, the volume differences will be reflected in the thickness of the organic phase. The results (Fig. 4) showed that the amount of extraction increased with decreased volume of organic phase. When the volume of organic solvent increased to 0.05 mL, the amount of extraction reduced to about 0.06 mg. The experimental results were consistent with the diffusion law that is the diffusion amount is inversely proportional to the diffusion distance at certain concentration.

Close

Figure 4

The effects of the volume of the organic phase on the extraction quantity.

Export to PPT

3.1.4 Influence of pH value of the donor phase

Acidic aqueous solution is used as the acceptor phase when the analyte is alkaloid. Alkaloid molecules reached in the acceptor phase will combine with an H⁺ to form water soluble salt. Such salts are not soluble in the organic phase, therefore, cannot return to the donor phase. Hence, the acidity of acceptor phase has somehow influenced the extraction efficiency. Experiments were investigated under five different acidity accepter phases (calculated with 85% phosphoric acid volume fraction) at 45 °C when 0.2 mL chloroform used as the organic solvent and standard matrine solution (1000 mg L⁻¹) as the donor phase. Fig. 5 shows that the extraction amount increased as the acidity levels increased and achieved the largest when the acceptor phase was pure phosphoric acid. The amount of extracted matrine reached 0.95 mg after 2 h extraction. The study indicated that the acidity of the acceptor phase was one of the driving forces of membrane dialysis.

Close

Figure 5

The effects of pH value of the donor phase on the extraction quantity.

Export to PPT