|Home | About | Journals | Submit | Contact Us | Français|
A new series of organic-high ionic strength aqueous two-phase solvents systems was designed for separation of highly polar compounds by spiral high-speed counter-current chromatography. A total of 21 solvent systems composed of 1-butanol-ethanol-saturated ammonium sulfate-water at various volume ratios are arranged according to an increasing order of polarity. Selection of the two-phase solvent system for a single compound or a multiple sample mixture can be achieved by two steps of partition coefficient measurements using a graphic method. The capability of the method is demonstrated by optimization of partition coefficient for seven highly polar samples including tartrazine (K=0.77), tryptophan (K=1.00), methyl green (K= 0.93), tyrosine (0.81), metanephrine (K=0.89), tyramine (K=0.98), and normetanephrine (K=0.96). Three sulfonic acid components in D&C Green No. 8 were successfully separated by HSCCC using the graphic selection of the two-phase solvent system.
High-speed counter-current chromatography (HSCCC) [1–6] requires tedious work for selecting the suitable two-phase solvent systems by measuring the partition coefficients (K) of target compounds. It often consumes 90% of researchers’ work force only for optimizing the K values . In order to solve this problem, methods of systematical search for the suitable solvent systems have been devised for separation of moderately polar compounds. These solvent series include hexane-ethyl acetate--methanol or ethanol-water system [8, 9], and methyl tert.-butyl ether-acetonitrile-1-butanol-water system each at various volume ratios, the latter of which is conveniently used for pH-zone-refining counter-current chromatography . However, there are many biologically active polar compounds which are still not easily separated by the conventional HSCCC. Recently, we have introduced a series of polar organic-aqueous two-phase solvent systems composed of i-butanol-ethanol-saturated ammonium sulfate-water at various volume ratios which can be applied for separation of highly polar compounds using spiral column counter-current chromatography . This new solvent system series consisted of 10 steps with a gradual change of polarity so that the search can be performed by following the series up or down until the suitable K values for the target compounds are obtained. However, when the log K values of 4 polar samples were plotted against the solvent system numbers, it was found that each curve formed a near straight line. This trend suggested that the K values of each component in all solvent systems could be determined simply by plotting the K values of two solvent systems on a graph paper.
The present paper introduces a new series of polar solvent systems consisting of 21 steps from the same set of solvents. In this paper we devised a simple graphic method for selecting a two-phase solvent system which gives suitable K values for a single target compound or a mixture of multiple compounds. Using this graphic method an ideal two-phase solvent system can be found by two steps of K measurements. The utility of the system is demonstrated by several test samples with high polarity including amino acids, catecholamines, sulfonic acids, etc.
The HSCCC instrument used in the present studies is a type-J coil planet centrifuge (Ito Multi-Layer Coil Separator-Extractor, P.C. Inc, Potomac, MD, USA). It holds a separation column and a counter-weight symmetrically at a distance of 10 cm from the central axis of the apparatus. The spiral tube support for holding the multi-layer spiral column was purchased from CCBiotech, Rockville, MD, USA. The spiral column was prepared in our laboratory inserting a single piece of PTFE (polytetrafluoro-ethylene) tubing (Zeus industrial Products, Orangeburg, SC, USA) after extruding it through a narrow slit followed by twisting along its axis (flat-twisted spiral column) . The column consisted about 50 m of 1.6 mm ID PTFE tubing with a total capacity of 70 ml. It has 10 layers of spiral tubing each containing 4 interwoven spiral tubing connected in series. The revolution speed of the apparatus can be regulated up to 1200 rpm with a speed controller, but the present experiments were performed at 800 rpm. The solvent was pumped with a HPLC pump (Model: LC-10ADvp, Shimadzu Corp., Kyoto, Japan) and the effluent was monitored with a UV detector (Model: Uvicord SII, LKB Instruments, Bromma, Sweden). The chromatogram was recorded with a strip-chart recorder (Model: Pharmacia LKB-Rec 102, Sweden). The partition coefficient values of test samples were determined with a spectrophotometer (Model: Genesys 10 UV, Thermo Spectronic, Rochester, NY, USA).
Absolute ethanol of analytical grade was purchased from Warner Graham Co., Cockeysville, MD, USA, 1-butanol of HPLC grade from Fisher Scientific Co., Fair Lawn, NJ, USA, and ammonium sulfate of analytical grade from Mallinckrodt Chemicals, Phillipsburg, NJ, USA. Test samples including methyl green (Matheson Coleman and Bell, Cincinnati, OH, USA), L-tyrosine (United States Biochemical Corp., Cleveland, OH, USA), DL-tryptophan (Janssen Chimica, Geel, Belgium), tyramine hydrochloride (Aldrich, Milwaukee, WI, USA), DL-normetanephrine (Sigma, St. Louis, MO, USA), DL-metanephrine (Sigma) were purchased from the companies given in the parentheses. The sulfonated dyes, D&C Yellow No. 5 (tartrazine) and D&C Green No. 8 (pyranine), were research samples provided by Dr. A. Weisz of U.S. Food and Drug Administration, College Park, MD. The samples of 1,3,6-pyrenetrisulfonate (P3S) and 8-hydroxy-1,3,6-pyrenetrisulfonate (HP3S) were prepared previously , and the sample of 1,3,6,8-pyrenetetrasulfonate (~98%, monohydrate, P4S) was obtained from Sigma-Aldrich, St. Louis, MO, USA.
First, ammonium sulfate saturated solution was prepared as follows: To a large glass bottle containing water placed on a magnetic mixer, ammonium sulfate was added until the solution became cloudy. The mixing was continued for several hours until the solution returned to room temperature. The bottle was kept for overnight to yield clear supernatant which was used for the preparation of two-phase solvent systems as described below. This solution contains ca. 4 M of ammonium sulfate with density at 1.260 g/cm3.
In the present study a set of 21 two-phase solvent systems from systems 0 to 20 were prepared as listed in Table 1. Each solvent system was made by mixing 1-butanol, ethanol, saturated ammonium sulfate solution and water at the indicated volume ratios each in a 50 ml capacity container. The contents were well mixed to equilibrate the two phases. After preparation, the physical properties of the solvent system such as a volume ratio of the two phases (upper/lower), density difference between the two phases and the settling time were determined. The settling time of the two phases was measured according to the standard technique  as follows: 2 ml of each phase (total 4 ml) of the equilibrated solvent system was delivered into a test tube (13 × 100 mm). After capping the test tube, the contents were vigorously shaken for 5 times and left vertically on the test tube rack to measure the time required to form clear two layers. The test was repeated 3 times to get the average value.
Partition coefficient values of the test samples in the selected two-phase solvent systems were determined using a spectrophotometer. In each measurement a small amount of the sample was added to 4 ml of two-phase solvent system (2 ml each phase) in a test tube. After capping the test tube, the contents were vigorously stirred with a vortex mixer until the sample was completely dissolved and partitioned into the two phases. After two phases were separated, an aliquot (usually 100 μl) of each phase was separately mixed with 2 ml of solvent (water for the lower phase and 50% ethanol for the upper phase) to measure the absorbance at a suitable wavelength using a spectrophotometer. The K values are expressed by the absorbance value of the upper phase divided by that of the lower phase.
Separation of sulfonic dyes was performed with an optimized two-phase solvent system composed of 1-butanol-ethanol-saturated ammonium sulfate-water at a volume ratio of 1.2:0.8:1:1 (System 8). The experiment was initiated by filling the column entirely with the upper organic phase followed by sample injection. Then the column was rotated at 800 rpm while the lower aqueous mobile phase was pumped into the internal head terminal of the spiral column at a flow rate of 0.5 ml/min. The effluent from the outlet of the column was continuously monitored with a UV detector at 280 nm and fractionated into test tubes at 4 minute intervals. After all peaks were eluted, the column contents were pushed out by pressured air into a graduated cylinder to measure the volume of the stationary phase retained in the column.
Table 1 illustrates compositions of a set of 21 two-phase solvent systems which are arranged from the top to the bottom in an order of increasing polarity. Important physical properties of the solvent system such as a volume ratio between the two phases, settling time, and density difference between the two phases are also presented. All systems show acceptable volume ratios between the two phases. The settling times of all these systems are longer that 20 seconds suggesting that the conventional multilayer coil may not be efficiently used for separation. However, the spiral tube configuration used in the present studies can retain a satisfactory volume of the stationary phase at over 50% by utilizing the centrifugal force gradients acting along the spiral channel. A large difference in density between the two phases due to a high concentration of ammonium sulfate in the lower phase also enhances the retention of the stationary phase in the spiral column. A set of these polar solvent systems can be used to determine the suitable two-phase solvent systems for a single compound or a multiple sample mixture by only two steps of K measurements using a simple graphic analysis as described in the following sections. Although lower phase fraction contains a large amount of ammonium sulfate, it is largely removed by adding methanol followed by centrifugation.
A suitable K value for separation of a single target compound is near unity. In order to achieve this goal, one may often spend an enormous amount of time to optimize the solvent system. However, the ideal two-phase solvent system can be easily found from the solvent series listed in Table 1 by two-step K measurements followed by a simple graphic analysis as follows: For each sample prepare three two-phase solvent systems, Systems 0, 10 and 20 in Table 1. Then, measure the partition coefficient of the sample in System 10 to get K10. If the sample is a pure standard as in the present study, K value may be determined with a spectrophotometer at a suitable wavelength. If the sample is a crude mixture, the standard HPLC method should be used to get the K value of the target compound. Then if K10 is greater than 1, the second K measurement is performed in System 0 to get K0. If K10 is above 1, the second measurement is performed in System 20 to get K20. Then, these two Log K values are plotted in the graph as shown in Figure 1. Then, these two points were connected with a straight line to get the crossing point to the Log K = 0 line which indicates the ideal solvent system for the separation of this compound. Several compounds including tartrazine, methyl green, tryptophan, and catecholamines have been tested, the results of which are shown in Table 2. It should be noted that all these compounds are not efficiently separated by the standard HSCCC method with existing two-phase solvent systems due to their low K values.
The present method can also be applied to the selection of the two-phase solvent system for separation of multiple compounds as follows: The sample mixture is first distributed in System 10. If the averaged K values are greater than 1, the second measurement is performed in System 0. But if the averaged K value is smaller than 1, the second measurement is performed in System 20. In either case, a pair of log K values for each target compound is plotted on the graph and connected with a straight line. In Fig. 2, a set of three sulfonic acids, P3S, HP3S and P4S were each plotted according to this procedure. From this graph the separation of these three compounds is easily visualized by distribution of three lines at each solvent system. In this case System 8 or 9 will provide satisfactory peak resolution between three compounds. Fig. 3 shows the separation of D&C Green No. 8 which contains three compounds at a flow rate of 1 ml/min at 800 rpm.
In order to expand the utility of HSCCC, a novel series of organic-high ionic strength aqueous two-solvent systems have been designed. It consists of 1-butanol-ethanol-saturated ammonium sulfate-water at various volume ratios which are arranged according to the order of increasing polarity. Using a graphic analysis, the suitable two-phase solvent system for a single component or multiple sample mixture is determined by 2-step partition coefficient measurement. The capability of the method is demonstrated by separation of sulfonic dyes.
This work was supported by the Intramural Research Program (IRP) of the National Institutes of Health (NIH). The work also was partially supported by projects (No. 09ZA036, KFJJ0905/0807, and 2011JQ0032) of Sichuan Province and National Natural Science Foundation of China (81101101).