Free Books and Brochures

and it becomes even more difficult and expensive for columns of wider diameter. There are other problems that need addressing, once packed, the practical lifetime of a column is also uncertain. The changes in performance of a preparative HPLC column that occurs with time depends upon the stability of the packed bed. Frequently, the bed settles after operation for even a short time and the top of the column needs to be repacked. Sometimes channels are formed in the bed, in which case the entire column has to be repacked. The rate of settling again depends upon the diameter of the column. This bed instability arises because there is a significant change in wall support as the column diameter increases. In analytical columns the walls are relatively close to the center of the column and 'bridges' of packing particles can be formed across the bed, as shown in Figure 16. These bridges allow the longitudinal forces acting on the packing within the column to be dissipated to the walls. When

through� column (1) to port (5), from port (5) to port (4) and out to the detector. Thus, the separation will take place in column (1). The ports connected to column (2) are themselves connected by the third slot and thus isolated. When the valve is rotated, the situation is depicted on the right hand side of figure (7); port (1) is connected to port (2), port (3) connected to port (4) and port (5) connected to port (6). This results in the mobile phase from either a sample valve or another column entering port (1) passing to port (2) through column (2) to port (3), then to port (4) and then to the detector. The ports (5) and (6) are connected, this time isolating column (1). This arrangement allows either one of two columns to be selected for an analysis or part of the eluent from another column pass to column (1) for separation and the rest passed to column (2). This system, although increasing the complexity of the column system renders the chromatographic process far more

the film thickness is also determined by the physical properties of the surface, the solvent and the stationary phase. The coating procedure is depicted in figure 15. After the plug has been run into the front of the column (sufficient to fill about 10% of the column length), pressure is applied to the front of the column to force the plug through the column at 2-4 mm per second (it will take about 5.5 hours for the plug to pass through a 60 m column). When the plug has passed through the column, the gas flow is continued for about an hour. The gas flow must not be increased too soon, or the stationary phase solution on the walls of the tube is displaced forward in the form of ripples, which produces a very uneven film. After an hour the flow rate can be increased and the column stripped of solvent. The last traces of the solvent are removed by heating the column above the boiling point of the solvent at an increased gas flow rate. Complete solvent removal can be

and the connection of the column inlet to the sampling system is now directed to waste. At the same time the column exit is disconnected from the detector and connected to a separate carrier gas supply that forces a backward flow of carrier gas through the column and the strongly retained solutes are eluted to waste. As suggested above, to accelerate the purging process, the column temperature can be raised. When back flushing is complete, the valve is returned to the sampling position and the column temperature brought back to the initial conditions for analysis. The sampling stage of the back flushing technique shown in figure 20 and depicts the sample passing through the valve to the column and from the column back to the valve and through the detector. The normal split injection system places the sample on the column.   Figure 21. Removal of Highly Retained Solutes in the Back Flushing Procedure After the solutes of interest have been eluted, the valve is rotated, and the

in figure 10.     Figure 10. The Dynamic Coating Procedure for an Open Tubular Column   The procedure is as follows. A plug of stationary phase solution is run into the front of the column (sufficient solution should be added to fill about 10% of the column length) and then connected to a gas supply. Pressure is applied to the front of the column and adjusted so that the plug velocity through the column is about 3 mm/second. After the plug has passed through and out of the column, the gas flow is continued for about an hour. The gas flow must not be increased too rapidly after the plug has left the column otherwise the film 0f stationary phase solution at the walls of the column will be displaced forward in the form of ripples and consequently produce a very uneven film. After an hour, the flow rate can then be increased and the column stripped of solvent by evaporation. This procedure requires some experience if evenly coated columns of the desired film thickness

may need to be determined by experiment. During the initial pressure adjustment some of the packing passes into the column and forms a lightly packed bed at the bottom of the column. The exit valve is hen rapidly opened and the sudden flow of gas packs and compacts the bed at the same time. After packing, the reservoir is carefully removed so as not to loosen the top of the packing and connected to the sampling system. LC Columns If particle sizes in excess of 20 mm are used, then the column can often be dry packed, with appropriate tapping, or, even better, with longitudinal and radial sonic vibration. The variance per unit length obtainable from a preparative LC column should be less than 2 particle diameters (determined using analytical scale samples). It is worth remembering that (as already discussed) when designing preparative columns, it is better to obtain the necessary efficiency using a longer column packed with larger particles, than the converse. The long column