Paper No. : 01Fundamentals of Analytical Chemistry Module :24 Chromatographic methods II
Principal Investigator: Dr.NutanKaushik, Senior Fellow
The Energy and Resouurces Institute (TERI), New Delhi Co-Principal Investigator: Dr. Mohammad Amir, Professor of Pharm. Chemistry,
JamiaHamdard University, New Delhi
Paper Coordinator: Prof. Rajeev Jain, Professor of Chemistry, Jiwaji University, Gwalior
Content Writer: Prof. Rajeev Jain, Professor of Chemistry, Jiwaji University, Gwalior
Content Reviwer: Dr. NimishaJadon, Assistant Professor, Jiwaji University, Gwalior
Paper: Fundamentals of Analytical Chemistry
Module: 24
Chromatographic Methods II
INTRODUCTION
The common features of the kinds of chromatography to be discussed in this lecture are:
(a) a column packed with a porous stationary phase, and (b) a liquid mobile phase to elute the sample components through the column.
Description of Module
Subject Name Analytical Chemistry / Instrumentation Paper Name Fundamentals of Analytical Chemistry Module Name/Title Chromatographic methods II
Module Id 24
Pre-requisites Objectives Keywords
Liquid chromatography is particularly useful for samples containing large molecules or ionic substances with low vapor pressures, and for thermally unstable substances which cannot be vaporized without decomposing. Liquid eluents are seldom inert; therefore, the distribution coefficients depend on the chemical nature of both the stationary and the mobile phase. Liquids have much higher viscosities and greater resistance to flow than do gases. In addition rates of diffusion of solutes are five orders of magnitude slower in liquids than in gases; thus, the C term in the Van Deemter equation is much larger than the A and B terms at normal flow rates. Van Deemter plots for gas and liquid chromatography are compared in Figure 1. Liquid chromatography is a time consuming procedure, because low flow rates are indicated for best efficiency.
Figure 1 Typical Van Deemter plots for gas and liquid chromatography
LIQUID - LIQUID PARTITION CHROMATOGRAPHY
Partition chromatography offers great advantages over adsorption chromatography. It is far more reproducible and predictable from solubility data. Even more important, the distribution coefficient is constant over a much greater range of concentration, yielding sharper, more symmetrically shaped bands. It is the method of choice whenever a suitable solvent pair can be found.
Some care must be given to the choice of solvent pairs. They must, of course, be immiscible. The more polar of the two is normally coated on a solid support. Optimally, the fraction of a solute which is dissolved in the mobile phase should be in the range from 0.05 to 0.5; otherwise the retention time will be too long or too short. Most often, a thin film of water supported on silica gel is used as the stationary phase. Up to 50% water can be adsorbed by silica gel before it becomes too moist to handle easily. For some purposes, it may be desirable to include a buffer system in the aqueous phase to change and/or control the solubility of the solutes.
Reverse -Phase Chromatography. There are some situations in partition chromatography
in which it is advantageous to use the nonpolar organic solvent as the stationary phase in
order to get a more favorable value for the partition coefficient. Obviously, a nonpolar solid
support is required. Powdered rubber coated with benzene is a very satisfactory stationary
phase with an aqueous mobile phase.
HIGH PERFORMANCE LIQUID CHROMATOGRAPHY
In liquid chromatography, maximum efficiency is achieved at low flow rates. This is due low rates of diffusion which are characteristic of liquid phases. The alternative is to reduce the distance through which the molecules must diffuse.
Packing Materials.
Closely sized particles as small as 10 µm are now becoming available. A solid glass bead of 30 to 50 µm in diameter can be coated with a layer (thin skin of 1 to 2 f-tm thickness) of porous material. These coated beads are called pellicular beads. The porous layer may serve as a solid stationary phase or be coated with a very thin layer of liquid stationary phase with an extremely large surface area.
High Pressures. Columns packed with small particles require high inlet pressures in
order to give a reasonable flow rate. Pressures up to 10,000 psi are not difficult to handle in the small columns used (2 to 3 mm diam.). Improved, pulse-free pumping systems are incorporated in modern liquid chromatographs.
Detectors. Flow-through detectors with low dead volumes and high sensitivity are a
necessity.
Two types of detectors are currently popular. One is based on a flow-through micro-cell placed in an ultraviolet spectrophotometer. A low pressure mercury lamp is commonly used as a source and the absorption measured with the 254-nm line where most organic compounds having double bonds or aromatic groups cause at least some absorption.
Measurements are not restricted to 254 nm if one employs a standard spectrophotometer.
Since most common solvents do not absorb radiation in the ultraviolet and visible region, these detectors are extremely sensitive and free from interferences.
A second popular detector, the differential refractometer, continuously monitors the difference in refractive index between the pure mobile phase (reference stream) and the column effluent. These devices respond to essentially all sample components, but the refractive index differences are minute and fluctuate with small changes in temperature or composition of the mobile phase (gradient elution is difficult).
Another detector utilizes the flame ionization detector developed for gas chromatography.
A continuous wire or fine chain moves through the stream emerging from the column outlet
where it becomes coated with a film of the effluent. The wire then passes through an oven
to evaporate the mobile phase, and finally through an oxidation oven in which the
components are burned in a stream of oxygen at about 850°C. The gaseous products are
passed over a nickel catalyst bed at 350°C which converts the carbon dioxide to methane, a
substance that is readily measured by the flame ionization detector.
Many variations have been tried in an effort to overcome the difficulties of the transport system; however, these devices offer the unique advantage of responding directly to the number of carbon atoms in the component molecule. A judicious combination of these improvements has led to separations of closely related compounds in very short times, such as the impressive separation of five hydroxylated aromatics in less than 60 sec described in Figure.
Figure 2: High speed separation of hydroxylated aromatics by liquid-solid chromatography.
Gas chromatography BASIC GLC APPARATUS
Unlike a liquid chromatographic system, a gas chromatograph requires a completely closed system (except for the gas outlet at the end). The essential components are shown in Figure 3. The carrier gas, supplied from a pressurized tank, passes through one or more pressure regulators which control the flow rate through the apparatus. The sample is introduced into a heated chamber either through a silicone rubber septum with hypodermic syringe if it is a liquid, or by means of a special sampling valve if it is a gas. From here the carrier gas carries the sample components through the column where they are separated and one after the other pass through a detector which sends a signal to a recorder. A thermostatted oven is provided for the column, injector, and detector, although the last two may be heated separately.
There are innumerable variations of each of these basic components of a gas chromatograph, and scores of commercial instruments are available in a wide price range for various specific applications. Whereas one pH meter is much like another, gas chromatographs should be selected on the basis of their applicability to the problem at hand, their versatility, accuracy, ease of service, and price.
Figure 3 Essential features of a gas chromatograph.
CARRIER GAS
The choice of carrier gas is usually based on availability in a high purity grade, or on the requirements of the detector which must sense the component in an extremely high dilution of carrier gas. For example, thermal conductivity detectors work best with hydrogen or helium. For most purposes, helium is the popular choice.
Flow through the column is caused by the difference in pressure between inlet and outlet.
Pressure regulating valves maintain an inlet pressure, Pi, of 10 to 50 psig (pounds per square inch, gauge-above atmospheric) or occasionally higher. The outlet pressure, Po, is normally atmospheric pressure although it can be increased by restricting the outlet orifice.
SAMPLE INTRODUCTION
As in other types of chromatography it is important to introduce the sample in the shortest time and in the smallest volume possible. The sample chamber may be heated for rapid vaporization of liquid samples and should be designed so that the carrier gas sweeps the sample directly into the column. If these requirements are not met, the sample will be unnecessarily spread out before the separation process has begun. In some instruments, the sample is injected directly into the column at the inlet (on-column injection). This is preferable for most samples unless they have a very high boiling point.
The ordinary laboratory chromatograph can handle liquid samples in the range of 0.1 to 10 µL and gaseous samples in the range from 1 to 10 mL. A capillary column (see below) requires much smaller samples, of the order of 10-3 to 10-2 µL. Samples of this size must be introduced by a splitting technique, as indicated in Figure 4. In this way only a small fraction of the injected sample is used, the remainder is vented to the atmosphere. Accurate measurement of samples of this size is questionable and may account for much of the error in quantitative analysis. For reliable ther- modynamic data, retention values should be obtained with several sample sizes and extrapolated to zero sample (infinite dilution).
Figure 4 Splitter for obtaining very small samples.
Liquid samples are injected from a micro syringe of appropriate capacity through a rubber septum. A swift, neat motion of the plunger is necessary. Gas samples are easier to measure accurately because of their larger volume. For routine gas analysis, a two-position, hexport valve, shown in Figure 5, is convenient and accurate. Solid samples can be dissolved in a solvent, or sealed in a thin-walled glass vial which is then inserted into the sample chamber and finally crushed from the outside.
Figure 5 Schematic of typical hexport sampling valve: (a) carrier gas is bypassed directly to column, sample gas feeds through sample loop; (b) carrier gas is diverted through sample loop, sample gas is bypassed to vent.
COLUMNS
There are two distinct types of columns in common use, packed and open tubular (capillary). The packed columns are easier to fabricate, less expensive, last longer, have a higher capacity, and suffice for all but the most difficult separations. The open tubular columns have less pressure drop and therefore can be made much longer (more plates).
Packed columns are usually 1 to 20 m long and 3 to 10 mm 0/8 in. to 3/8 in.) in diameter, although columns up to 10 cm or more in diameter are used for preparative scale work. Open tubular columns are usually 10 to 50 m long and 0.2 to 1.2 mm in diameter. The geometrical configuration of the column is of little concern as long as the bends are not sharp. Columns are bent in a U- or W-shape or coiled to fit the oven. Short columns are often made of glass, but longer columns are made of copper, aluminum, or stainless steel tubing to facilitate bending which may be done after the column is filled. It is possible to draw glass capillary tubing in lengths of several hundred meters-capillary columns up to a mile in length have been fabricated.
COLUMN PERFORMANCE
Successful resolution of complex mixtures often requires careful attention to column efficiency which is indicated by n, the number of theoretical plates in the column, or by H, the height equivalent of a theoretical plate. In a more expanded version, the Van Deemter equation for a packed column is given as
(1) A term B term C1 term Cg term
where λ, γ and ω are functions of the packing structure. The A and B terms are as given previously but the C term has been divided into two parts: Cl (liquid phase) and Cg (gas phase)-both of which contribute to the mass transfer effect. In the C term, the thickness of the liquid film appears as d2f. Note that H will be different for each solute, because Cl depends on k' (and therefore on K). Remember that k' is determined by the choice of liquid phase, the temperature, and the ratio of the volumes of the two phases, k' = KVL/VG• Ds is the diffusion coefficient of the solute in the stationary (liquid) phase. The Cg
term is often ignored with packed columns because of the short distances that solute molecules have to travel in the gas phase in order to reach the stationary phase and the much larger value of the diffusion coefficient in the mobile phase, DM, compared to Ds. However, its effect becomes relatively more important with very thin liquid films. The particle diameter, dp, appears in both the A term and the Cg
term; in fact, Giddings has proposed that these two effects (non-uniform paths and cross-column transport in the mobile phase) are interactive, or coupled, adding another term to Equation 1. The effect of the coupling is that a large A term tends to decrease the effect of the Cg term.
SOLID SUPPORT
The ideal solid support is not yet available. It should have a high specific surface (l m2/g), and the surface must be chemically inert although wettable by the liquid phase so that the latter will spread in a thin layer of uniform thickness. In addition, the solid support must have thermal stability, mechanical strength, and be available in uniformly sized, near-spherically shaped particles.
The most commonly used supports are derived from diatomaceous earth, a spongy siliceous material consisting of the skeletal residues of diatoms, a form of microscopic algae. (Diatomaceous earth is also used as a filter aid, as the main constituent of firebrick, and as a mild abrasive.) The raw material has a surface area of about 20 m2/g, but it is very fragile and, of course, loaded with impurities.
In one process the diatomaceous earth is mixed with clay and baked at 900°C. It is then crushed and graded according to size. The final material has a surface area of about 4 m2/g. The surface is fairly active toward polar compounds and is pink; hence, designations like Chromosorb-P. In a second process, the raw material is mixed with a sodium carbonate flux before baking (calcining). The product is more fragile, less chemically active, has a smaller surface area (0.5 to 1 m2/g), and is white; hence designations like Chromosorb-W.
These support materials are interlaced with a network of fine pores, requiring about 0.5% (by weight) of a liquid to cover the entire surface with a monolayer. As a heavier coating is applied, the
finer pores fill up first, but even with a 20% coating many free passageways still exist, and the average thickness of the layer is still only a few hundred angstroms.
The surface of commercial firebrick (Chromosorb-P) has the general structure:
The -OH groups are acidic and somewhat polar, and tend to react with polar solutes, especially those with basic functional groups. The surface activity can be partially reduced and some impurities removed by washing in acids or bases. For the separation of amines, it is desirable to leave a thin coating of sodium hydroxide on the surface. A very effective treatment consists of silanizing the surface with hexamethyldisilazane (HMDS):
In this way the polar -O-H is replaced by a relatively inert trimethyl silyl group.
Powdered Teflon is useful as a support for very polar solutes, but it has a low surface area.
It is so inert that it is difficult to coat evenly. It is more easily handled if first cooled in a refrigerator. Some other supports which have been suggested are micro glass beads, graphitized carbon, and carborundum.
Coating the Support. The column packing is prepared by mixing the solid with the correct amount of liquid phase dissolved in a suitable low-boiling solvent such as pentane,
dichloromethane, or acetone. The solvent is then evaporated with judicious heating; stirring as necessary to obtain a uniform coating. The last traces of solvent may be removed under vacuum. Columns are usually filled by pouring the packing into the straight tube with gentle shaking or tapping. Both ends are plugged with glass wool and the column bent to the appropriate shape to fit the oven. Alternatively, bent columns can be packed by loading the packing material in a high velocity gas stream. Great care is necessary in packing large columns to avoid channeling and segregation of the particles according to size.
Open tubular columns, of course, contain no packing. The thin coat of liquid phase is applied by forcing a dilute solution through the column at a slow rate. The solution remaining on the wall is evaporated by passing through carrier gas, leaving a layer of liquid phase.
Experience helps in getting a satisfactory coating.
LIQUID PHASE
The versatility of GLC is in large part due to the wide variety of liquid phases available.
Hundreds of liquids have been tried, but perhaps a dozen or so will suffice for most purposes.
The requirements for a good liquid phase are: (1) it should be essentially nonvolatile (vapor pressure <0.1 torr) at the temperature it is to be used; (2) it must be thermally stable; (3) it should be readily available in a reproducible form. No single liquid meets all requirements for all possible solutes.
Some qualitative concepts are given below which will help us to understand the factors which determine retention at the molecular level.
1. Dipole-Dipole Interactions. Molecules containing electronegative or electropositive atoms possess a permanent electrical dipole, and will interact strongly with other
molecules possessing a dipole. Thus a polar solute will have an abnormally low volatility (high solubility) in a polar solvent. On the other hand, in a nonpolar solvent, the dipole- dipole interactions between the polar solute molecules themselves are decreased by dilution, resulting in an abnormally high volatility (low solubility). Dipole-dipole interactions are decreased as the temperature is increased.
2 Induction Forces. If either the solvent or solute contains a permanent dipole, it can induce a temporary dipole in the other. The magnitude of the force depends on the polarizability of the other molecule and is generally rather small.
3 Dispersion Forces. The vibrations of nonpolar molecules often produce temporary dipoles by slight separation of the electrical charges within the molecules. The oscillating dipoles can induce similar temporary dipoles in neighboring polarizable molecules. A small force of attraction is thus generated. These forces are present in all solutions.
4 Hydrogen Bonds. An especially strong dipole-dipole interaction is possible when one molecule contains a polarized hydrogen atom and the other a strong electronegative atom, such as a fluorine or oxygen atom. The extra strength results from the closeness of approach afforded by the small size of the proton.
5 Formation of Metal Complexes. Solutions of silver nitrate in glycols or benzyl cyanide selectively absorb olefins because of the weak organometallic complexes formed. Thus the olefins are retained far longer than corresponding paraffins in these columns.
Similarly, heavy metal salts of fatty acids retard amines because of complex formation.
DETECTORS
The remarkable separations performed in the column must somehow be sensed and recorded. All of the components are highly diluted in the carrier gas with concentrations of 1
part per thousand at best, ranging down to zero. Furthermore, sharp peaks may pass through the detector in less than a second, while the last peaks may not emerge for hours and be barely discernible above the base line. Somehow the detector must ignore the large amount of carrier gas and find the trace amounts of sample components contained therein.
Types of Detectors. Signals from differential detectors are usually integrated for quantitative
analysis, and signals from integral detectors are often differentiated to make them easier to interpret for qualitative analysis. Either signal in Figure 6 can be derived from the other.
Figure 6 Differential and integral response for the same chromatogram.
Detectors can also be classified as destructive or nondestructive, depending on whether or not the sample components can be collected unchanged for further study.
Hydrogen Flame Detector. One of the simplest detectors is the hydrogen flame detector, shown in Figure 7 with an exquisitely simple chromatograph. Hydrogen must be used as a carrier gas. It is burned as it emerges from the column through a hollow needle, yielding a nearly colorless flame. When an
organic component emerges, the flame becomes yellow. The retention time can be measured with a stopwatch. The amount of the component is roughly proportional to the height and/or luminosity of the flame. This may sound very crude, yet it is not difficult to install a thermocouple to measure the flame temperature or a photocell to measure the luminosity and increase the accuracy many fold.
Better still, since most organic compounds are ionized in the flame, an ion current can be collected between two oppositely charged electrodes, as in Figure 8. This is the principle of the flame ionization detector, one of the most sensitive and popular detectors in current use. The ions produced are collected between two electrodes, one of which may be the jet itself. Because the electrical resistance of the flame is very high (about 1012 ohms) and the current is extremely small (about 10-10 amp), the associated electronics are complicated and moderately expensive.
Figure 7 Simple chromatograph with hydrogen flame detector.
Figure 8 Hydrogen flame ionization detector
The ionization processes occurring in the flame are not completely understood. The ion current, however, is approximately proportional to the number of carbon atoms entering the flame. The detector is insensitive to most inorganic compounds. It is especially worth noting that it does not "see"
water vapor or air. The flame ionization detector is relatively simple, extremely sensitive, and has a wide range of linear response.
Thermal Conductivity Detector. The measurement of the thermal conductivity of a gas is based on
the transfer of heat from a hot filament to a cooler surface. Thus the gas conducts heat from the filament to the wall. If a constant amount of electrical energy is supplied to the filament, its temperature will be a function of the thermal conductivity of the gas. Rather than determining the temperature of the filament, it is easier to determine its electrical resistance which increases with temperature. As applied to gas chromatography, a dual detector is used to minimize the effect of the thermal conductivity of the carrier gas and to minimize fluctuations in the temperature, pressure, and power supply.
A schematic diagram of the detector is shown in Figure 9, and the associated electrical circuit in Figure 10. In some detectors, thermistors are used in place of filaments. Thermistors are somewhat more sensitive that filaments below about 100°C, but the reverse is true above about 150°C.
The temperature of the filament (and thus the signal) is a function of the bridge current, the geometry of the cell, and the thermal conductivity and flow rate of the gas. Increasing the bridge current will increase the running temperature of the filaments and increase the sensitivity, but the filaments will burn out sooner. Since the detector is measuring thermal conductivity, it is very important to keep the temperature of the detector walls constant; often this is the limiting factor in quantitative analysis.
Figure 9 Thermal conductivity detector
Figure 10 Circuitry for thermal conductivity detector.
The thermal conductivity of a mixture of gases is not easy to calculate, and this detector must be calibrated for each compound for highest accuracy. Some generalizations have been observed; for example, the relative response (signal per mole of compound/signal per mole of standard, in this case, benzene) for many compounds follows the relation
Relative Response = A + BM (benzene = 100) (2)
where M is the molecular weight, and A and B are constants for a given homologous series of compounds, a series in which a member is derived from the previous member by the insertion of a CHz
group.
Many forms of geometry of the cell cavities have been proposed to reduce the effects of flow fluctuations without decreasing the sensitivity or increasing the time constant unduly. The internal volume of the detector, and especially the dead space, should be at a minimum in order not to remix the components. The thermal conductivity detector is simple, rugged, inexpensive, moderately sensitive, non- selective, essentially nondestructive, very accurate if properly calibrated, and more widely used than any other.
Electron Capture Detector. In this detector, the effluent gas is ionized by a stream of particles
emanating from a radioactive source, typically 3H or 63Ni. Thus, the carrier gas produces a steady supply of positive ions and free electrons which can be measured as a standing current between two charged electrodes, much like the hydrogen flame detector. When sample components pass through the detector, the standing current may be perturbed. Compounds containing highly electronegative atoms capture free electrons very efficiently, and they are detected by the decrease in the standing current. This occurs because of the increased rate of recombination of positive and negative ions, compared with positive ions and electrons. Other species have little effect on the ion current, thus the
electron capture detector is highly selective and very sensitive for compounds containing halogens, phosphorus, lead, nitro groups, and polynuclear aromatic ring systems. It is ideally suited to the detection of minute traces of many pesticides and defoliants.
Other Detectors. Of the many other detectors proposed, some of the more important are: (a) the
gas density balance for which the response is a precise function of molecular weight; (b) coulometric titrator in which the column effluent is burned to give HCl, H2S, etc., and then passed through a solution to be titrated with electrolytically generated silver ion (specific for halogens and sulfur); (c) thermionic detector (sensitized flame ionization detector) of which there are several versions, each incorporating a FID provided with a screen or porous block coated with an alkali halide at or just above the flame.
(Depending on which salt is used, these detectors are highly selective and several orders of magnitude more sensitive to phosphorus-, sulfur-, or halogen-containing compounds); and (d) flame photometer detector in which the flame of an FID serves as a radiation source of a flame photometer. (It is also selective and sensitive to phosphorous-, sulfur-, and chlorine-compounds.)
V. Do you know
1. The detector must ignore the large amount of carrier gas and find the trace amounts of sample components contained therein.
2. Improved, pulse-free pumping systems are incorporated in modern liquid chromatographs.
3. The size of the sample is dictated by several factors: the amount available, the capacity of the column, and the sensitivity of the detector.
VI. Interesting facts:
1.
Molecules containing electronegative or electropositive atoms possess a permanent electrical dipole, and will interact strongly with other molecules possessing a dipole.
2.
The temperature of the filament (and thus the signal) is a function of the bridge current, the geometry of the cell, and the thermal conductivity and flow rate of the gas.
3.
At ordinary pressures and temperatures, the common carrier gases are regarded as chemically inert.
4.
Flow through the column is caused by the difference in pressure between inlet and outlet.
Analytical Chemistry / Instrumentation
Fundamentals of Analytical Chemistry Chromatographic methods II
