Column Selectivity Testing

An area of intense investigation is choosing the column with the best selectivity. The selectivity is dependent on the bonded phase (i.e., bonded ligand, silanol activity), the probe analytes, the pH of the mobile phase, the type and 

To ascertain if the selectivity between a set of probe compounds is reproducible from column to column with the same stationary phase, analysis on different columns should be performed. For example, three different columns from different lots of base silica should be used as well as three columns from the same lot of base silica. This will test the batch-to-batch reproducibility and the intra-lot reproducibility. At least 10 injections should be made on each column with the different probe mixes, and the selectivity across the columns from injection to injection should be compared. 

Therefore, a system suitability test with a known set of probes such as MIX 1 could be used as an internal test to provide further conhdence in regard to the batch-to-batch reproducibility of the packing material and/or to observe if the bonded phase has been compromised. This could also be used to probe the lot-to-lot reproducibility of new types of stationary phases that are available on the market. Once this simple and fast system suitability test is performed with MIX 1 and acceptable results are obtained using a set of defined acceptance criteria, the analysts may commence with his/her analytical method and run the specihc system suitability test stated in the method for their particular target pharmaceutical analyte. 

---

SRM 869a Column Selectivity Test Mixture for Liquid Chromatography [44] is composed of three shape-constrained PAHs (phenanthro[3,4-c]phenanthrene, PhPh l,2 3,4 5,6 7,8-tetrabenzonaphthalene, TBN and benzo[a]pyrene, BaP) and is routinely employed to evaluate the shape selectivity of stationary phases. The retention differences between the nonplanar TBN and planar BaP solutes (expressed as a selectivity factor axEN/BaP = provide a numerical assessment of... 

Certificate of Analysis, Standard Reference Material 869a column selectivity test mixture for liquid chromatography (polycyclic aromatic hydrocarbons), National Institute of Standards and Technology (NIST), Gaithersburg, MD, 1998. Available at http //www.nist.gov/SRM... 

GL 1] [R 1] [R 3] [P le] The performance of a typical laboratory bubble column was tested and benchmarked against the micro reactors (Figure 5.17). Using acetonitrile as solvent, the conversion of the laboratory bubble column ranged from 6 to 34% at selectivities of 17-50% [3, 38]. This corresponds to yields of 2-8%. Hence the yields of the laboratory tool are lower than those of the micro reactors, mainly as a consequence of lower selectivities. 

It is necessary to use the test methods presented in this section with some caution. Not all of them have been adequately standardized or applied in a broad sense. They should prove useful for aiding column selection and improving the properties of column packings prepared in the laboratory. It should also be noted that columns may deteriorate with use depending on their treatment. An old coltum of a type that has been shown to perform well in a particular test in the literature may not live up to expectations, due to its altered state. 

Each of the ( ) column selections must be tested. Any systematic way of ordering these selections so that none are omitted is acceptable. This additional procedure to avoid repetitions is carried out in the last part of Example 9 in Section VI. 

For the characterization of the selected test area it is necessary to investigate whether there is significant variation of heavy metal levels within this area. Univariate analysis of variance is used analogously to homogeneity characterization of solids [DANZER and MARX, 1979]. Since potential interactions of the effects between rows (horizontal lines) and columns (vertical lines in the raster screen) are unimportant to the problem of local inhomogeneity as a whole, the model with fixed effects is used for the two-way classification with simple filling. The basic equation of the model, the mathematical fundamentals of which are formulated, e.g., in [WEBER, 1986 LOHSE et al., 1986] (see also Sections 2.3 and 3.3.9), is ... 

A column performance test is used to ensure the proper condition of columns and the stability of retention parameters. Because CWC-related chemicals greatly differ both chemically and physically from each other, the test chemicals have been selected so that their physical, chemical, and retention properties are different, and so that they elute evenly over the whole chromatogram. The use of the following chemicals in a column performance test is recommended trimethylphosphate, 2,6-dimethylphenol, 5-chloro-2-methylaniline, tri-n-butylphosphate, dibenzothiophene, malathion, and methyl stearate. The concentration of test chemicals depends on the sensitivity of detectors. The... 

The final part of the chapter provides a refresher on pA from an analytical chemist s perspective, the drivers for choosing normal phase versus reversed phase as a separation mode for a particular analysis, and instru-ment/system consideration, and it concludes with a very interesting section on column testing within the framework of bonded phase stability (effect of pH and type of buffer) and probing column selectivity. 

In Figure 8-66, the same selectivity test mix (MIX 1) and conditions were used for a polar embedded column. Multiple columns from three different lots... 

Smitz et al. [ 114] subjected to multivariate analysis the retention data of 9 selected test analytes determined on 26 reversed-phase columns. By PCA the columns were grouped in three classes of similar properties. The authors were able to detect the deviation of some columns from the typical trend observed for RP-8, RP-18 or polymer-coated materials. The deviation could be explained in terms of the extreme physicochemical properties of the column. 

The Prontosil column is very similar to, although not identical with, Genesis. It could be used as an alternative if the latter one is not available. On the other hand, the Platinum is very different and cannot be used as a substitute. However, it could be interesting to use it in order to check if all peaks of a mixture are separated in the course of selectivity tests during validation. 

Tosio et al. [70] suggested that in order to identify and determine acid components in mixtures with neutral compounds, a flow of carrier gas should be passed through a reactor (100 X 0.5 cm l.D.) packed with potassium hydroxide on a quartz powder (115 100) after the chromatographic separation of the initial mixture on an analytical column. Selective absorption of the acid components takes place in this reactor. By comparing the chromatogram obtained with the analytical column and that obtained with an alkaline reactor, it is possible to identify and determine the acid and neutral components of the test mixture. As an example, results were presented of the analysis of small amounts of... 

Many modern packed column SFCs are equipped with a column selection valve and a solvent selection valve. Software Wizards are available that allow simple programming where a large number of different stationary and mobile phases can be automatically tested to find the best conditions without operator intervention. 

The simplest form of an HPLC SST involves comparison of the chromatogram with a standard one, allowing comparison of the peak shape and the peak width baseline resolution. Additional parameters that can be experimentally calculated to provide quantitative SST report include the number of theoretical plates, separation factor, resolution, tailing or peak asymmetry factor, accuracy, and precision (RSD of six measurements). Resolution may also be combined with a selectivity test to check the resolution of the analytes from components present in the sample matrix. If matrix components interfere with a method, a matrix blank may be included in the SST. Peak shape and asymmetry, or tailing factor, can have a major impact on the performance of a quantitative method, especially at low amounts or concentrations, and may vary over the lifetime of a column. 

How well these selectivity groupings work, and how these similarities translate into the everyday separations of the chromatographer s work is left to the judgement of the reader. There are many separations in many laboratories that can be carried out on more than one stationary phase. The column selection used for the creation of the charts described here is rather large, and the reader will surely find at least some of the columns that he or she is commonly using. With the information provided here, the interested user can then test the described classification scheme and see how well it applies to his or her separation problem. 

Chromatographic Tests and Their Importance in Column Selection... 

First of all, the number of test compounds to be employed should be as low as possible. In our ongoing example, we show how a set of 14 test compounds can be successfully reduced to seven, while still maintaining the same quality of column evaluation. The 14 previously selected test compounds possessed different physicochemical properties and could be classified on the basis of size and polarity. 

Another important issue in the optimization of a chromatographic test is the selection of the most appropriate chromatographic conditions to analyze the test compounds. More specifically, depending on the mobile phase composition, different kinds of information on the chromatographic performance of stationary phases can be gleaned. In a first attempt, columns were tested with two mobile phases composed of pH 7.0 phosphate buffer (one with acetonitrile and the other with methanol) and one composed of a pH 3.0 phosphate buffer. 

Multivariate linear regression can be carried out with common statistics software and was performed in the given example with Statistica 5.0 (StatSoft). Retention factors were measured for a set of selected test solutes using a set of different columns under one fixed set of eluting conditions (acetonitrile/water, 3 7, vjv, without addition of a buffer). A buffer was avoided, since Engelhardt and coworkers reported a marked modulation of column properties by salt or buffer additives this adulterates phase characterization and makes columns look less silanophilic than they actually are [15]. 

The mathematical plan for determining the parameters from the chromatographic data is divided into eight steps, as discussed in the following. In a first step, the retention of all i selected test solutes is related to that of ethylbenzene (ideally hydrophobic substance by definition) according to Eq. (7). This results in a data point log oq p for each column P. 

So far, values of H, S, etc. that characterize column selectivity have been reported for about 200 columns [12-16]. Values for another hundred columns are included in a commercial database (Column Match Rheodyne LLC, Rohnert Park, CA). A procedure for measuring values of H, S, etc. has been reported and shown to give comparable values for the same columns in inter-laboratory testing [17]. Table 2 summarizes average values of H, S, etc. for several different types of columns. Within a given column type, further changes in these column parameters are typical, as illustrated in Table 1 for some type B Cjg columns. 

---