Chromatography

Background

Clinical tests often involve the use of one or more steps to isolate, enrich, or separate a target compound from other chemicals in the sample. Chromatography is one of the most common methods for achieving this type of separation. In this method, the components of a mixture are separated based on their differential interactions with two chemical or physical phases: a mobile phase and a stationary phase that is held in place by a supporting material. There are many forms of chromatography based on the different mobile phases, stationary phases, and supports that can be used in this method, which has led to a wide range of applications for this technique.

Content

This chapter describes the basic principles of chromatography and discusses various forms of this method that are used for chemical analysis or to prepare specimens for analysis by other techniques. The methods of gas chromatography and liquid chromatography are discussed, as well as the techniques of planar chromatography, supercritical fluid chromatography, and multidimensional separations. The mobile phases, stationary phases, and supports that are used in each of these methods are described. The instrumentation and detection schemes that are employed in these methods are also discussed.

Biological fluids are complex mixtures of chemicals. This means that clinical tests for specific components in these fluids often involve the use of one or more separation steps to isolate, enrich, or separate the target compound of interest from other chemicals in the sample. Chromatography is one of the most common methods for achieving this type of separation. This chapter describes the basic principles of chromatography and discusses various forms of this method that are used for chemical analysis or to prepare specimens for analysis by other techniques.

General terms and components of chromatography

Chromatography is a method in which the components of a mixture are separated based on their differential interactions with two chemical or physical phases: a mobile phase and a stationary phase. The basic components and operation of a typical chromatographic system are illustrated in Fig. 19.1 . The mobile phase travels through the system and carries sample components with it once the sample has been applied or injected. The stationary phase is held within the system by a support and does not move. As a sample’s components pass through this system, the components that have the strongest interactions with the stationary phase will be more highly retained by this phase and move through the system more slowly than components that have weaker interactions with the stationary phase and spend more time in the mobile phase. This leads to a difference in the rate of travel for these components and their separation as they move through the chromatographic system.

The type of chromatographic system that is shown in Fig. 19.1 uses a column (or a tube) to contain the stationary phase and support, while also allowing the mobile phase and sample to pass through the system. This approach was first described in 1903 by Mikhail Tswett, who used this method to separate plant pigments into colored bands by using a column that contained calcium carbonate as both the support and stationary phase. Tswett gave the name chromatography to this method. This name is derived from Greek words chroma and graphein, which mean “color” and “to write.” This term is still used to describe this technique, even though most modern chromatographic separations do not involve colored sample components.

The type of chromatography that was used by Tswett, in which the stationary phase and support are held within a column, is known as “column chromatography.” In chromatography, the stationary phase may be the surface of the support, a coating on this support, or a chemical layer that is cross-linked or bonded to the support. ^{, ,} In column chromatography, the support may be the interior wall of the column or it may be a material that is placed or packed into the column. A column is the most common format for chromatography. However, it is also possible to use a support and stationary phase that are present on a plane or open surface. This second format is known as “planar chromatography,” as will be discussed in more detail later in this chapter. ^,

One way of classifying chromatographic methods is based on the type of support that they employ; two examples are the techniques of column chromatography and planar chromatography. Chromatographic methods also can be classified based on the mobile phase that is present. For instance, a chromatographic method that uses a gas mobile phase is called gas chromatography (GC), and a chromatographic method that uses a liquid mobile phase is known as liquid chromatography (LC). It is also possible to divide chromatographic methods according to the type of stationary phase that is present or the way in which this stationary phase is interacting with sample components. Examples of these classifications include the GC methods of gas-solid chromatography (GSC) or gas-liquid chromatography (GLC) and the LC methods of adsorption chromatography, partition chromatography, or ion-exchange chromatography (IEC). Each of these categories, as well as others, will also be discussed later in this chapter.

The instrument that is used to perform a separation in chromatography is known as a chromatograph. ^, For instance, in GC the instrument is a gas chromatograph, and in LC the instrument used to carry out this method is a liquid chromatograph. These instruments can provide a response that is related to the amount of a compound that is exiting (or eluting) from the column as a function of the elution time or the volume of mobile phase that has passed through the system. The resulting plot of the response versus time or volume is known as a chromatogram, ^, as is illustrated in Figs. 19.1 and 19.2 .

The average time or volume that is required for a particular chemical to pass through the column is known as that chemical’s retention time ( t _R ) or retention volume ( V _R ). These values both increase with the strength and degree to which the chemical is interacting with the stationary phase. The elution time or volume for a compound that is nonretained or that does not interact with the stationary phase is known as the void time ( t _M ) or void volume ( V _M ). If the retention time or retention volume is corrected for the void time or void volume, the resulting measure of retention is known as the adjusted retention time ( t _R ′, where t _R ′ = t _R − t _M ) or the adjusted retention volume ( V _R ′, where V _R ′ = V _R − V _M ). For two chemicals to be separated by chromatography, it is necessary for these chemicals to have different values for t _R and V _R (or t _R ′ and V _R ′). ^{, ,}

Most separations that are used for chemical analysis in column chromatography are carried out by injecting a relatively small volume or amount of sample onto the chromatographic system. This situation results in a chromatogram that consists of a series of peaks that represent the different compounds in the sample as they each elute from the column. The retention time or retention volume of each peak can be used to help identify the eluting compound, whereas the area or height of the peak can be used to measure the amount of the compound that is present.

The width of each peak is also of interest in a chromatogram. The peak width reflects the separating performance or efficiency of the chromatographic system. The width of a peak in a chromatogram is often represented by its baseline width ( W _b ) or its half-height width ( W _h ) ( Fig. 19.3 ). ^{, ,} As the widths for the peaks in a chromatogram become sharper, it becomes easier for the chromatographic system to separate two peaks with similar interactions with the system and to separate more peaks in a given amount of time. Sharper peaks are also easier to measure than broader peaks and tend to produce better limits of detection.

Retention and selectivity

For two chemicals to be separated by chromatography, these chemicals need to have some differences in how they are interacting with the stationary phase versus the mobile phase. Besides using the retention time and retention volume (or adjusted retention time and adjusted retention volume) to describe these differences, another way of representing retention in chromatography is by using the retention factor ( k ). This term is also sometimes represented as k ′ or called the capacity factor. ^, The retention factor is a measure of the average time a chemical resides in the stationary phase versus the time it spends in the mobile phase. This value can be calculated from experimental data by using any of the following equivalent relationships ^{, ,} :

or

As these equations suggest, the retention factor is a unitless number where a value of 0 indicates that no binding or interactions are occurring between a chemical and the stationary phase or that this compound is eluting from the system at the void time. As the chemical undergoes greater interactions with the stationary phase, this will result in longer retention times and an increased value of k . In practice, it is desirable to have a value for k that is between 1 and 10 to provide reasonable separations between compounds without the need for excessive lengths of time for their elution from the column.

The retention factor is useful in describing a compound’s retention in chromatography for several reasons. First, the value of k should be independent of the flow rate and column size. Also, k can be directly related to the strength of the interactions that are occurring between a chemical and the stationary phase or mobile phase, as well as the relative amount of stationary phase versus mobile phase that is present in the column. This last feature is illustrated by the following equation for a chromatographic system in which a chemical is separated based on its ability to partition between the mobile phase and stationary phase. Similar relationships can be written for other types of separation mechanisms :

In this relationship, the value of k is directly related to (1) the distribution equilibrium constant ( K _D ) for partitioning of the analyte into the stationary phase versus the mobile phase and (2) the relative amount of stationary phase in the column (as represented here by V _S ) versus the amount of mobile phase that is present (as represented by V _M , the void volume). The value of k in this situation will increase if there is either an increase in K _D , which reflects the tendency of the chemical to enter the stationary phase over the mobile phase, or the ratio ( V _S / V _M ), which is a term also known as the phase ratio . ^,

Any separation in chromatography requires that there be some difference in retention for the chemicals that are to be separated from each other. One way of describing this difference in retention is by using the separation factor or selectivity factor (α). ^, The separation factor for two compounds (A and B) is equal to the ratio of their retention factors ( k _A and k _B ),

where the retention factor for the later eluting component is given in the numerator. If two chemicals have the same retention in a chromatographic system, the value of α will equal 1 and no separation will be possible. If the peaks for A and B do have different retention, the value of α will be greater than 1 and will increase as the degree of separation increases.

The values of both the retention factor and selectivity factor are determined by the chemicals that are being separated, as well as the stationary and mobile phases that are present in the chromatographic system. A large difference in retention and a large separation factor are desirable when the goal is to selectively isolate one chemical from others in a sample. However, smaller differences in retention and in separation factors are often used when the chromatographic system is used to separate several chemicals and peaks from the same sample. In this second situation, a value for α of 1.1 or greater represents an adequate separation in many common types of chromatography. However, chromatographic methods that result in broad peaks may need even larger values of α to produce a good separation between two chemicals.

Band-broadening and efficiency

Besides needing a difference in retention for a separation to occur, the peaks for two neighboring chemicals must be sufficiently narrow to allow this difference to be observed. The injection of even a sample with a small volume will experience some increase in width, or band-broadening, as this peak travels through the chromatographic system. This broadening of peaks is produced by various processes related to the rate of movement or diffusion of the applied chemicals as they pass around or within the support and within or between the mobile phase and stationary phase. These band-broadening processes, in turn, are affected by factors such as the diameter or type of support within the chromatographic system, the flow rate, the diffusion coefficient of the chemical in the mobile phase and stationary phases, and the degree of retention of the chemical in the column ( Box 19.1 ). Together, these processes and factors determine the overall efficiency or extent of band-broadening obtained.

The efficiency and degree of band-broadening in a chromatographic system are related experimentally to the final observed width of a chemical’s peak. This width can be described by measures such as the baseline width ( W _b ), the half-height width ( W _h ), or the standard deviation (σ) of the peak. These values, in turn, can be used to find another measure of chromatographic efficiency known as the number of theoretical plates, or plate number (N). The value of N for any type of chromatographic peak can be calculated by using the following formula,

where t _R is the retention time for the peak and σ is the standard deviation of the peak in the same units of time as t _R . ^, This equation takes on the following two equivalent forms for a Gaussian-shaped peak. ^,

These last two equations make use of the fact that a Gaussian peak has a baseline width, as measured by the intersection of the baseline with tangents along either side of the peak, that is equal to 4 σ, and a half-width width that is equal to 2.355 σ.

The value of N can be thought of as representing the effective number of times that a chemical has been distributed between the mobile phase and stationary phase as this chemical has passed through the chromatographic system. A larger value for N represents many such steps, which makes it easier to distinguish between two chemicals that have only small differences in their retention. Experimentally, a large value of N results in a high chromatographic efficiency and sharp peaks, which are both desirable for either separating chemicals with similar retention or quickly separating many chemicals in the same sample.

There are several other ways in which the efficiency of a chromatographic system can be described. One way is by using the number of theoretical plates (N) per unit length of the chromatographic system ( L ), as given by the ratio ( N/L ). This ratio helps in comparing systems with different lengths, because the value of N increases in direct proportion to the length of the column or support bed that is used in a separation for chromatography. Although this means that a longer chromatographic system will always lead to a larger value for N and greater efficiency, the use of a longer system also results in longer separation times.

Another way of describing column efficiency is the height equivalent of a theoretical plate or plate height (HETP, or H ). ^, The value of H is found by dividing the length of the chromatographic system by the number of theoretical plates for this system.

The value of H represents the length of the column or chromatographic system that makes up one theoretical plate or one distribution step for a chemical between the mobile phase and stationary phase. Although a large value of N (or N / L ) represents a chromatographic system with high efficiency, the same system would be represented by a small value for H (or L / N ).

A valuable feature of using H to describe chromatographic efficiency is that this term can be related directly to the parameters and processes that affect band-broadening. A common example of this is the van Deemter equation , which shows how the overall value of H is affected by the linear velocity of the mobile phase (u), which is directly related to the flow rate (F) through the relationship u = ( F × L )/ V _M . ^,

The terms A, B, and C in this equation are constants that represent the contributions of several types of band-broadening processes. For instance, the A term represents the contributions of band-broadening processes that are independent of the linear velocity and flow rate, such as eddy diffusion and mobile phase mass transfer. The B term is the contribution to the plate height by longitudinal diffusion, which is a process that becomes more important as the flow rate and linear velocity are decreased. Finally, the C term represents the contributions from processes that lead to an increase in H as the flow rate or linear velocity is increased. The processes that make up the C term are stagnant mobile phase mass transfer and stationary phase mass transfer. The van Deemter equation predicts that the combined effect of these band-broadening processes will be an optimum range of flow rates and linear velocities over which the lowest plate heights, and best efficiencies, will be obtained. In practice, the usual goal in varying the flow rate in chromatography is to identify those conditions that provide the most rapid separation times while still providing adequate resolution of all peaks that are of interest in the samples being separated.

Several factors that affect chromatographic efficiency are listed in Box 19.1 . For instance, efficiency can be improved by using longer columns, which increases the value of N but does not alter H . It is also possible to change the flow rate to its optimum value, to use smaller diameter support particles, to use nonporous or pellicular particles instead of fully porous support particles, or to use a relatively narrow-diameter coated capillary instead of a packed bed column. All these latter factors help to lower the value of H, which in turn increases the value of N for a given length of column or chromatographic bed. However, there are practical limits to how much some of these experimental parameters can be changed. As an example, a reduction in the diameter of the support particle will lead to greater efficiency, but it will also result in higher back pressures across the chromatographic system, require the use of lower flow rates, or both.

Resolution and peak capacity

The overall extent to which two peaks are separated in chromatography can be described by using a term known as the resolution (R _s ) , as is illustrated in Fig. 19.3 . The resolution between two neighboring peaks can be found by using the following formula ^, :

In this equation, t _R,A and t _R,B are the average retention times for compounds A and B (where B elutes after A), while W _b,A and W _b,B are the baseline widths for the peaks of these compounds (in time units, in this case). An equivalent equation can be written in terms of the retention volumes of A and B and their baseline widths in volume units. The use of either approach will give a unit-less value for R _s that represents the average number of baseline widths that separate the centers of the two peaks.

Fig. 19.4 shows how the separation of two neighboring peaks changes as the value of R _s increases for these peaks. An R _s value of 0 is obtained when there is no separation between the peaks and they have exactly the same retention times or retention volumes. The degree of peak separation increases as the value of R _s increases. An R _s value of 1.5 or greater is often said to represent a complete separation between two equally sized peaks, or baseline resolution. However, for many separations, resolution values between 1.0 or 1.25 and 1.5 also may be adequate, especially if the peaks are about the same size and are to be measured using their peak heights rather than their peak areas.

Several approaches can be used to alter or improve the resolution between two peaks in chromatography ( Fig. 19.5 ). These approaches are indicated by the following expression, which is sometimes known as the resolution equation of chromatography :

In this equation, k is the retention factor for the second of two neighboring peaks, α is the separation factor between the first and second peaks, and N is the number of theoretical plates for the chromatographic system. This relationship indicates that resolution of two peaks in chromatography can be changed in three ways: (1) by altering the efficiency of the system, as represented by N; (2) by changing the overall degree of peak retention, as represented by k; or (3) by changing the selectivity of the column for one compound versus another, as represented by α. An increase in N, such as can be obtained through use of a longer column, will lead to an increase in R _s that is proportional to N ^1/2 . An increase in the retention factor (k) or selectivity (α) will also lead to a nonlinear increase in resolution.

Another way of describing a chromatographic separation is in terms of the peak capacity . The peak capacity is the maximum number of peaks (or sample components) that can be separated, in theory, during a single chromatographic separation. The value of the peak capacity can be found by assuming there is a continuous distribution of peaks that are separated by an average baseline width (or 4 standard deviations). In practice, the number of components that can be separated in a single run by a given system will be lower than the theoretical peak capacity because the retention times of their peaks will not be evenly distributed. The peak capacity of a system that is used for high-performance liquid chromatography (HPLC) is usually limited to several hundred peaks, whereas higher values can be obtained in methods such as capillary GC. Factors that can be used to increase the peak capacity include increasing the efficiency of the system (e.g., by using a longer column) and using gradient elution or extended run times. Another approach for increasing the peak capacity is to use a multidimensional separation, as will be discussed later in this chapter.

Analyte identification and quantification

Chromatography is often used as an analytical tool to qualitatively identify analytes in a sample and to measure the concentrations of these analytes. For example, the retention time, retention volume, and retention factor for an analyte are all characteristic values that reflect how this chemical is interacting within a particular chromatographic system. These retention values can be compared to those for a known sample of the same compound to help confirm its identity. However, other confirmation also may be needed because other compounds may have similar retention characteristics.

One-way additional confirmation can be obtained is if the unknown compound and reference compound have the same retention under several types of chromatographic conditions, such as on different columns or column/mobile phase combinations. In the case of capillary GC or LC columns, it is possible to simultaneously introduce samples onto two columns that contain different stationary phases and that are connected to separate detectors. If the unknown compound and a reference compound match in their retention properties on the two columns, this greatly enhances the chance for correctly identifying the unknown analyte. An alternative and even more reliable approach for identification is to use a detection method that provides structural information on the analyte, such as mass spectrometry (see Chapter 20, Chapter 21, Chapter 22, Chapter 23 ).

The peak area or peak height can be used to produce quantitative information on an analyte that is separated from other sample components by chromatography. Peak areas tend to provide a more precise means for measuring an analyte, whereas peak heights are easier to use if there is not complete resolution between the analyte and its neighboring peaks. Both external and internal calibration techniques can be used in chromatography for such measurements. ^, In external calibration, standard solutions containing known quantities of the analytes are processed and separated in the same manner as samples that contain one or more of these analytes ( Fig. 19.6 ). A calibration curve is then constructed by plotting the peak height or peak area (or the spot density, in the case of planar methods) versus the concentration or mass of analyte that was applied in the standard solutions. This curve can then be used with the peak area or peak height that is determined for the same analyte in the samples to find the concentration or amount of this analyte present.

In the method of internal calibration (also called internal standardization), standard solutions of the analyte are again prepared; however, a constant amount of a different compound known as the internal standard is also now added to each standard solution and sample ( Fig. 19.7 ). The internal standard should be a chemical that was not originally present in either the sample or the standard, is similar in its chemical and physical properties to the analyte, and can be measured independently from the analyte. This internal standard is typically added to the samples and standards before they are processed by any pretreatment steps, such as extraction or derivatization. The addition of this agent can help normalize the results for any variations that may occur during the pretreatment steps or during sample/standard injection onto the chromatographic system. This normalization is made by constructing a calibration curve in which the y -axis is based on the ratio of the peak height or peak area for the analyte in a given standard or sample divided by the peak height or peak area for the internal standard in the same standard or sample. This ratio is plotted versus the concentration or amount of analyte in each standard. This calibration plot can then be used to find the concentration or amount of the analyte that was present in each sample. ^,

Types of gas chromatography

There are several ways of classifying GC methods based on the type of stationary phase present. These categories include GSC, GLC, and bonded phase GC.

Gas-liquid chromatography and bonded phases

In GLC, the stationary phase is a liquid that is placed as a coating or layer on the support. This is the most common type of GC for chemical analysis. Various types of liquids can be used for this purpose (see examples in Table 19.1 ). All these liquids must have a low volatility to allow them to stay within the column at the high temperatures that are often used in GC separations. Many GLC stationary phases are based on polysiloxanes, which have the basic structure shown in Fig. 19.8 . The molar mass of the –Si-O-Si- chain in a polysiloxane can range in size from a few thousand to over a million grams per mole. The side chains that are attached to the silicon atoms in this chain can have structures that range from nonpolar methyl groups to polar cyanopropyl groups. These side chains also can be present in various ratios as mixtures. The overall polarity and types of chemicals that will be retained the most by this type of stationary phase will be determined by the amounts and types of side chains that are present.

TABLE 19.1

Stationary Phases Commonly Used in Gas-Liquid Chromatography and as Bonded Phases in Gas Chromatography

Composition	Polarity	Commercial Examples	Typical Applications
100% Methylpolysiloxane	Nonpolar	OV-1, SE-30	Drugs, amino acid derivatives
5% Phenyl–95% methylpolysiloxane	Nonpolar	OV-23, SE-54	Drugs
50% Phenyl–50% methylpolysiloxane	Intermediate polarity	OV-17	Drugs, steroids, glycols
50% Cyanopropylmethyl–50% phenylmethylpolysiloxane	Intermediate polarity	OV-225	Fatty acid methyl esters, carbohydrate derivatives
Polyethylene glycol	Polar	Carbowax 20M	Acids, alcohols, glycols, ketones

One issue in using a liquid as a stationary phase in GC is that some of this liquid will eventually leave the column over time. This loss of the stationary phase is known as column bleed. ^, This process is not desirable because it will result in a change in the amount of stationary phase present and a change in the ability of the GC system to retain chemicals. This process also may cause the signal of the GC detector to have a high background or to be noisy as the liquid stationary phase leaves the column and passes through the detector.

Column bleed can be minimized by using a bonded phase instead of a liquid as the stationary phase in the GC column. The resulting method is sometimes known as bonded phase GC. A bonded phase can be produced by reacting functional groups on a stationary phase such as a polysiloxane with silanol groups on the surface of silica. Alternatively, the stationary phase can be cross-linked to make it less volatile and more stable. Besides providing a stationary phase that is more stable, a bonded phase also can provide a stationary phase that has a thinner and more uniform coating than a stationary phase based on a liquid coating. Although bonded phases are more expensive than liquid stationary phases, bonded phases are often preferred for analytical work because of their better thermal stability and better efficiencies. ^,

POINTS TO REMEMBER

Types of gas chromatography based on the stationary phase

• Gas-solid chromatography

• Gas-liquid chromatography

• Bonded phase gas chromatography

Gas chromatography instrumentation

The typical components of a gas chromatograph are illustrated in Fig. 19.9 . The first major component is the source of the gaseous mobile phase, which is used to supply the carrier gas at a controlled pressure and flow rate. Next, there is the injection system, through which samples are placed into the gas chromatograph and converted into a volatile form. This is followed by the column, which contains the support and the stationary phase. This column is held in an oven for temperature control. The fourth part of the GC system is a detector that monitors sample components as they leave the column. Finally, there is a computer or control system that acquires data from the detector and allows control of the GC system.