Monitoring Column Performance

In general, new columns should be calibrated in the laboratory in which they will be used, and a standard mobile phase and a standard series of compounds should be available for this purpose. The resolution obtained under the standard conditions of the laboratory at the start of the useful life of the column should be recorded, together with the date of the analysis. All should be part of the record for that column. At the first sign of problems with this column, its performance should be checked against these records, using the same mobile phase and standards. 

Chromatography involves the separation of classes or groups of molecules. Two phases are usually required one is stationary or solid, and the other mobile—eluent or buffer. 

If the solid phase is in the form of particles or beads, it is usually packed into a tube or column and the buffer or mobile phase is pulled through the packing by gravity or forced through with a pump. 

The time required for a given compound to emerge from the column is a function of the packing material and the interactions between the compound and the packing. Transition time is affected by the distance the compound travels or the rate at which it travels a fixed distance. 

The basic equipment required for an HPLC system includes a solvent reservoir, a pump, an injector, an analytical column, a detector, and a recorder. The analysis of the sample is displayed as a chromatogram, with detector deflection presented usually as a function of time after loading the sample. By virtue of the shape of the curves, the distance between them, and their 

---

Additional temperature indicating or recording points should be included up the column for monitoring column performance and for trouble shooting. 

The aids to chromatography include a) resolution calculations on chromatograms of standard mixtures to monitor column performance, b) calculation of Kovats retention index for help in identifying peaks, and (c) multiple point calibration curves for improved quantitation. The file searching routines access two sets of data. Information (such as molecular formula, molecular weight) is stored on 3100 compounds from the Arctander data( ). This allows a quick computer search through the data which is difficult... 

The problem of column maintenance, particularly when the column is used for enzyme assays, is discussed, cleaning solutions are recommended, and a method for monitoring column performance described. 

As is quite evident from interpreting the ehromatograms in Fig. 4.50, RP-HPLC columns have a finite lifetime It is good practice for the analyst to keep a record of N (as calculated using the above equation) versus either time or number of injected samples in an attempt to continuously monitor column performance. Because HPLC reciprocating pumps maintain constant flow rate, a continuous observation of the back-pressure or pressure buildup at the front on the HPLC column is an important parameter to monitor. Making sure that there are no leaks in an operating HPLC is also very important. 

Table 12. Suggested standard test procedures for monitoring column performance.

It is common for the test mixture used by a manufacturer to check the performance and quality of an HPLC column to be totally umelated to the assay for which the column was purchased because they are made up of quite simple compounds, which are readily available in the laboratory. When setting up a test procedure for monitoring column performance over its lifetime, the same procedure as the manufacturer should be used if possible. It is useful to note that a different result for the efficiency of a column will occur for each different test method, and it is therefore important for each HPLC laboratory to standarize on one dy. Suggested test mixtures, along with their respective test conditions, are outlined in Table 12. Commonly, two or three components are used, one with a very short retention time [typically with a k of about 0.2, where k is the column capacity factor (see Figure 19) to assess band broadening caused by... 

FIGURE 9.11 For each PSS SEC column, packing pressure and permeability are monitored for the best reproducibility of column performance. 

It is difficult to decide what should serve as adequate column quality parameters for describing the performance of a set of GPC columns. The two most common measures are plate count and resolution. While both of these can be useful for monitoring the performance of a column set over time, it is not generally possible to a priori specify the performance needed for a specific analysis. This will depend on the nature of the polymer itself, as well as the other matrix components. 

In general, chromatography columns should not be left connected to pumps or to the UV monitors in salt solutions. Always include a wash step with water to remove any salt from the system. It is preferable to store the columns disconnected in 20% ethanol and to rinse the entire FPLC system, including pumps, tubing, and UV flow cell with water, followed by 20% ethanol. Keep a record of the column performance, and use it to determine when filter changes or columncleaning steps are required. 

The longer a component is retained by the column, the greater is the capacity factor. To monitor the performance of a particular column, it is good practice to periodically measure the capacity factor of a standard, the number of plates (Equation 23-28), and peak asymmetry (Figure 23-13). Changes in these parameters indicate degradation of the column. 

The organic acids were analyzed with an HPLC system consisting of degassing unit (Gastorr GT-103),pump (Pharmacia LKB), autosampler (Pharmacia, LKB), precolumn (Shodex R Spak KC-LG), separation column (Shodex R Spak KC-811), water bath (Julabo U3) and a UV detector (Soma Optics LTD S-3702). The data monitoring was performed using a PC-chromatography data system Andromeda 1.6 (Techlab). 

Keeping records of column backpressure and important chromatographic parameters [number of theoretical plates (AO, peak asymmetry factor (As), retention factor k ), resolution factor (Rs)] helps to monitor the required column performance, while storage in the appropriate organic solvent extends column lifetime. Table 8 presents the preventive actions for column protection. 

Thixotropic and dilatant behavior of the suspension can be monitored by viscosi-metric measurements as a function of shear rate and the duration of the experiments. Studies with a 25 wt.% suspension of LiChrospher 100, 7 gm in a mixture of dioxane-cyclohexane-n-octanol (45/45/10, v/v/v) have shown that the suspension has rather low thixotropic and pseudo-elastic behavior (Hallmann, 1992). The packing pressure of the axial dynamic column packing technique has a significant influence on the column performance (Tab. 3.13). 

Dynamic simulation may be used for off-line or on-line applications. An offline dynamic model runs independent of the plant. This is a predictive mode in which a column or an entire process is simulated to predict transient behavior with no input from the plant. Such a model is typically used for the design of equipment and control strategies and as a training simulator. An on-line dynamic model may be used to monitor the column performance and to provide vital information for the control system. It reads current plant conditions and, in real time, computes properties that are not measured on-line, such as product compositions. This makes it possible to control such properties directly. The on-line dynamic simulator can also predict future column trends, thereby allowing the control system to take corrective action in advance. 

In order to complete the model, additional calculations at subsequent time intervals should be carried out. The calculations defined above reflect initial conditions at time interval 1. This should be followed by calculations for time intervals 2, 3,. .. Justification for this procedure is the fact that the process is unsteady state. The analysis at each time interval shows the change in phase conditions along the column height, at a given time. Subsequent time interval calculations represent time progression, where the column performance can be observed as a function of elapsed time. This allows monitoring the column conditions stage-wise and time-wise. 

---