Basic Concepts of High-performance Liquid Chromatography

The two basic questions in high-performance liquid chromatography focus on (a) how particular compounds can be separated, and (b) why particular compounds were separated by the liquid chromatographic method used. The answers can be obtained by the consideration of some simple representative chromatograms of the separation of well-known compounds. Such separations can be easily understood according to common principles of physics and chemistry.

A separation is described by the following equation, which indicates the degree of resolution between two peaks in a chromatogram, Rs. A complete separation requires Rs > 1.2 units.

Rs = 1/4(α - 1/α)[square root of N]

The resolution can be improved by increasing the column plate number, N, and/ or the separation factor, α(α = the ratio of the retention factors of the two compounds). N is the physical parameter and α is the chemical parameter for the separation. Higher N and α values give a better separation.

The physical and chemical aspects of liquid chromatography, in addition to mechanical aspects, are briefly described in this chapter. Theoretical approaches are explained in detail in later chapters. The effect of stationary phase materials on the chemical selectivity is described in Chapter 3, and the influence of the eluent components is covered in Chapter 4. The plate number theory is discussed in Chapter 5. Quantitative optimization is explained in Chapter 6.

1 Physical Parameters for High-speed Separations

It was thought that high-speed separations would be achieved by the development of a physically stable pumping system and highly sensitive detectors; however, the main contribution to high-speed separation is made by small-size stationary phase materials. A shorter separation time with complete resolution cannot be achieved simply by increasing the flow rate or by using a small column. The theoretical plate number of a small column must be the same as that of a larger column to obtain the same separation.

For example, the separation of a mixture of benzene, acetophenone, toluene, and naphthalene has been completed within 5.5 min using a 15 cm long, 4.6 mm i.d. column, packed with 10 µm porous octadecyl-bonded silica gel, whose theoretical plate number was 38 000 m-1, as shown in Figure 1.1 A. Increasing the flow rate 4-fold reduced the separation time to 1.5 min, because this mixture was well separated (Figure 1.1B). The same mixture was separated within 4.5 min using a 10 cm long, 4.6 mm i.d. column packed with 3 µm octadecyl-bonded porous silica gel with a theoretical plate number of 117 000 m-1 (Figure 1.1C). Doubling the flow rate resulted in completion of the separation within 2 min, as shown (Figure 1.1D).

Comparison of these four chromatograms suggests that a fast separation can be performed either using a longer column with 10 µm stationary phase material with a high flow rate of the eluent, to give high resolution, or by a smaller column packed with 3 µm stationary phase material. However, a high flow rate through the 3 µm stationary phase material is limited by a high column back pressure. The separation could also be completed within 1.2 min on the short column packed with 3 µm stationary phase material by using a stronger eluent, as shown in Figure 1.2. Furthermore, the sensitivity was also improved by using the smaller-size stationary phase material because the sample is less spread out in the eluent and is more concentrated when it reaches the detector. The actual peak height in Figure 1.1C is 1.6 times that in Figure 1.1 A. A small column packed with small particle-size stationary phase materials promises high performance and a high-speed separation both in theory and in practice. The following equation describes the relationship of the column length (L) to the column efficiency: N = L/H. The high plate number N required for good separation is proportional to the longer column length L and small H value. The term H is the height equivalent to a theoretical plate (HETP), which is the length of column needed to generate one theoretical plate. A good column has a high plate number for its length, and, thus, a good column has a low H value. The value of H can also be described by the following equation (which is described in detail in Chapter 5):

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]

This equation indicates that the particle size, dp, is the main contributor to the H value. The smaller the particles, the higher the theoretical plate number. The optimum condition is obtained by the relationship between the theoretical plate height and the flow velocity.

2 Physical Considerations

High-speed separations can be achieved with a short column packed with 3 µm stationary phase material, as shown in Figure 1.2. The sensitivity was also improved by the use of smaller-size stationary phase materials, due to less sample diffusion inside the column. The following conditions are required to obtain such a separation.

a. small-diameter, spherical stationary phase materials that have high physical strength;

b. a high pressure pump with controlled flow rate;

c. a system that limits sample diffusion, by considering the column design, using small inner diameter connecting tubing, and a small volume detector flow cell; and

d. a detector and recorder capable of a high-speed response.

The theoretical plate number N of peak B can be calculated from the chromatogram given in Figure 1.3 by the following equation:

N = 16(VR/w)2

where VR is the retention volume and w is the peak width at the base (measured in volume units). However, the retention volume includes the hold-up volume VM (also called dead volume). The hold-up volume is the sum of the void volume of the column (V0 = YA), the volume of the injector (OX) and the volume of the detector and connecting tubing (XY) as shown in Figure 1.3. The actual separation efficiency is defined as the effective theoretical plate number Neff, which excludes the hold-up volume:

Neff = 16(VR - VM/w)2

Commercial instruments have a reasonable balance between the recommended column size and the volume of the column and connecting tubing (XY). However, the theoretical plate number of a single column may give different values on different instruments, and even on replacement of the components and parts of a single instrument. Such discrepancies can be understood in terms of differences in the mechanics of the instruments and the design of their parts.

The normally acceptable extra-column dead volume (OY in Figure 1.3) before there is a significant effect on the efficiency of a 15 cm long, 4.6 mm i.d. column should be less than 100 µl. This volume has to be reduced to less than 30 µl for a 5 cm long, 4.6 mm i.d. column. Replacement of the connecting tubing with shorter lengths of narrow-bore tubing and the selection of a smaller volume detector flow cell are necessary when using a shorter or narrower column. These changes together with a smaller column enable reductions in the volume of eluent required and in the separation time. This approach is economical and environmentally friendly. However, the reduced hold-up volume becomes technically very critical in the handling of smaller columns. Details of the basic mechanisms and the design of instrumentation are described in Chapter 2, which also covers the similarities and differences of various instruments.

3 Chemical Influences on the Separation Factor

One column can be used for different types of liquid chromatography by changing the eluent components. As an example, a column packed with octadecyl-bonded silica gel has been used for size-exclusion liquid chromatography with tetrahydrofuran (THF), normal-phase liquid chromatography with n-hexane, and reversed-phase liquid chromatography with aqueous acetonitrile. Examples of the chromatograms are shown in Figure 1.4.

The elution volumes of polystyrene and benzene in the size-exclusion mode were 0.98 and 1.78 ml, respectively (Figure 1.4A). This means that separations by molecular size can be achieved between 0.98 and 1.78 ml in this system. In the normal phase mode the elution volumes of octylbenzene and benzene were 1.98 and 2.08 ml, respectively, in n-hexane solution (Figure 1.4B). This type of chromatography is called adsorption or non-aqueous reversed-phase liquid chromatography. These are adsorption liquid chromatography and non-aqueous reversed-phase liquid chromatography. The elution order of the alkylbenzenes in the reversed-phase mode using acetonitrile was reversed (Figure 1.4C). The elution volumes of benzene and butylbenzene were 2.01 and 2.52 ml, respectively. The elution volumes became larger with the addition of water to the acetonitrile eluent. In each case the elution orders are based on the solubility of the solutes (except in size-exclusion liquid chromatography).

When separation cannot be achieved by improving the theoretical plate number of a column, it may be achieved by the selection of an appropriate stationary phase material and/or eluent. The degree of separation, the separation factor α, is the difference in retention volumes of analytes. The separation factor of two compounds is given by:

α = VR2 - VM/VR1 - VM

where VR2i.e. an increase in the difference in retention volumes, can be achieved by using a different stationary phase material and/or eluent. Details of the selection of stationary phase materials and eluent are described in Chapters 3 and 4.

4 Basic Considerations of Liquid Chromatography

Identifying the most suitable separation conditions is the main objective of separation scientists. It is easier for skilled chromatographers, but this is a complicated subject for beginners. One approach is to find a chromatogram that exhibits the separation of a similar mixture. However, similar mixtures may have been separated under very different conditions; either the separation columns and/or the components of the eluent may have been different. Furthermore, the elution order is sometimes reversed. When an appropriate chromatogram is found in the literature, the conditions may need to be modified to take into account the other compounds in the mixture, any necessity for sample pre-treatment and the purpose of the separation. The chromatographer should be familiar with the capabilities and requirements of the following methods.

a. Pre-treatment of samples, stationary phase materials, and elution solvents;

b. Separations based on molecular size;

(i) aqueous size-exclusion liquid chromatography;

(ii) non-aqueous size-exclusion liquid chromatography;

c. Normal-phase liquid chromatography;

d. Reversed-phase liquid chromatography;

e. Ion-exchange liquid chromatography;

f. Ion-pair liquid chromatography;

g. Chiral separation and affinity liquid chromatography.

Choosing the sample pre-treatment method is difficult. The most important consideration is the final condition of the target compounds. What kind of solution is obtained? The type of solvent and the concentration of the target compounds are very important in the selection of the separation conditions.

The pre-treatment of stationary phase materials is also important for silica gel and ion-exchangers, even when a new column is being used. Pre-treatment of the elution solvent is also important. High-performance liquid chromatography grade solvents from different manufacturers contain different amounts of impurities. The purity of the water is especially critical. The specified solvent for a special preparation stage is often not compatible with the desired chromatography. A large amount of such a solvent should first be injected, followed by measurement of the background of the chromatogram.

How are analytes retained on, or in, a stationary phase? This depends on the physicochemical interaction between the analytes and the stationary phase material. When a strong solvent, in which the solute readily dissolves, is used for elution the solute is eluted very quickly from the column. The forces holding an analyte on the stationary phase are similar to those responsible for dissolution in the solvent. Eight solubility properties are recognized: van der Waals (London dispersion) forces, dipole–dipole, ion–dipole, Coulombic and repulsion forces, charge-transfer complexation, and hydrogen-bonding and coordination bonds. The molecular interactions that are probably involved in retention in liquid chromatography can be explained by these interaction properties and are summarized in Table 1.1. The retention of a particular molecule is not due to only one property, but rather to a combination of several properties. The probable interactions can be estimated from the chemical structure of the analytes and stationary phase materials.

The separation conditions employed for size-exclusion liquid chromatography are simple. A strong solvent for analytes and a suitable stationary phase material are necessary. If the impurities have high relative molecular masses (Mr), size-exclusion chromatography can be used effectively. Size-exclusion liquid chromatography in conjunction with a recycling system can also separate isomers; however, it is time-consuming and the columns are usually expensive. If a mixture of molecules with a Mr, of less than 2000 has to be separated and a recycling method seems to be insufficient for the separation, the following chromatographic technique can be carried out. If the sample concentration is large enough for chromatographic analysis, the eluted solution obtained by a size-exclusion chromatographic pre-treatment can be directly injected onto a liquid chromatograph using a syringe, after membrane filtration. If a good combination of stationary phase material and solvent cannot be found, then methods c–g in the above list are applied.

In reversed-phase liquid chromatography, increasing the molecular size increases the hydrophobicity of solutes and results in a greater retention volume. This indicates that the van der Waals volume is an important property in optimization. Increasing the number of substituents with π-electrons and hydrogen bonding increases the solubility in water, that is they increase the polarity of the solutes. This indicates that dipole–dipole and hydrogen-bonding interactions contribute to hydrophobicity. Therefore, these properties are important in controlling the retention volume in reversed-phase liquid chromatography. However, the π-electrons of stationary phase materials such as polystyrene gel and the hydrogen-bonding of non-endcapped bonded silica gels also contribute to the retention.

Many compounds can be analysed by methods c or d, and sometimes both methods c and d are employed. For a preparative-scale separation, method c (normal-phase chromatography) is suitable due to the easy removal of solvent. Identifying the best separation conditions for ion-exchange liquid chromatography (method e) is a little more difficult. Therefore, if the compounds can be separated using methods c or d, these methods should be used. Even saccharides, organic acids, and nucleic acids are often separated by methods c and d. The separation speed in ion-exchange liquid chromatography is also slower than that in normal and reversed-phase liquid chromatography.

Affinity liquid chromatography and chiral separations (enantiomer separations) require similar analyte properties. The solutes may have interactions through hydrogen-bonding, ligand formation, or Coulombic forces with the surface of stationary phase materials or the sites of additives; however, the selectivity is controlled by the steric effects of the structures of the analyte molecules and the recognition molecules (chiral selectors).

The physical and chemical properties of stationary phase materials are described in Chapter 3 (including methods for their synthesis) to clarify the differences in similar stationary phase materials supplied from different manufacturers. A detailed selection guide to solvents is given in Chapter 4. The unlimited selection of eluent components and their concentrations is a powerful force in developing separations in liquid chromatography. Although this area seems rather complicated, it is easy to understand the selection of a suitable eluent when you first identify the molecular properties of the analytes and solvents.