Gas Chromatography (GC or GLC) is a commonly used analytic technique in many research and industrial laboratories for quality control as well as identification and quantitation of compounds in a mixture. GC is also a frequently used technique in many environmental and forensic laboratories because it allows for the detection of very small quantities. A broad variety of samples can be analyzed as long as the compounds are sufficiently thermally stable and volatile.
How does gas chromatography work?
Like for all other chromatographic techniques, a mobile and a stationary phase are required for this technique. The mobile phase (=carrier gas) is comprised of an inert gas i.e., helium, argon, or nitrogen. The stationary phase consists of a packed column where the packing or solid support itself acts as stationary phase, or is coated with the liquid stationary phase (=high boiling polymer). Most analytical gas chromatographs use capillary columns, where the stationary phase coats the walls of a small-diameter tube directly (i.e., 0.25 m film in a 0.32 mm tube).
The separation of compounds is based on the different strengths of interaction of the compounds with the stationary phase (“like-dissolves-like”-rule). The stronger the interaction is, the longer the compound interacts with the stationary phase, and the more time it takes to migrate through the column (=longer retention time). In the example above, compound X interacts stronger with the stationary phase, and therefore lacks behind compound O in its movement through the column. As a result, compound O has a much shorter retention time than compound X.
Which factors influence the separation of the components?
1. Boiling point
The boiling point of a compound is often related to its polarity (see also polarity chapter). The lower the boiling point is, the shorter retention time usually is because the compound will spent more time in the gas phase. That is one of the main reasons why low boiling solvents (i.e., diethyl ether, dichloromethane) are used as solvents to dissolve the sample. The temperature of the column does not have to be above the boiling point because every compound has a non-zero vapor pressure at any given temperature, even solids. That is the reason why we can smell compounds like camphor (0.065 mmHg/25 oC), isoborneol (0.0035 mmHg/25 oC), naphthalene (0.084 mmHg/25 oC), etc. However, their vapor pressures are fairly low compared to liquids (i.e., water (24 mmHg/25 oC), ethyl acetate (95 mmHg/25 oC), diethyl ether (520 mmHg/25 oC)).
2. The polarity of components versus the polarity of stationary phase on column
If the polarity of the stationary phase and compound are similar, the retention time increases because the compound interacts stronger with the stationary phase. As a result, polar compounds have long retention times on polar stationary phases and shorter retention times on non-polar columns using the same temperature. Chiral stationary phases that are based on amino acid derivatives, cyclodextrins and chiral silanes are capable of separating enantiomers because one enantiomer interacts slightly stronger than the other one with the stationary phase, often due to steric effects or other very specific interactions. For instance, a cyclodextrin column is used in the determination of the enantiomeric excess in the chiral epoxidation experiment (Chem 30CL).
3. Column temperature
A excessively high column temperature results in very short retention time but also in a very poor separation because all components mainly stay in the gas phase. However, in order for the separation to occur the components need to be able to interact with the stationary phase. If the compound does not interact with the stationary phase, the retention time will decrease. At the same time, the quality of the separation deteriorates, because the differences in retention times are not as pronounced anymore. The best separations are usually observed for temperature gradients, because the differences in polarity and in boiling points are used here (for examples see the end of the chapter)
4. Carrier gas flow rate
A high flow rate reduces retention times, but a poor separation would be observed as well. Like above, the components have very little time to interact with the stationary phase and are just being pushed through the column.
5. Column length
A longer column generally improves the separation. The trade-off is that the retention time increases proportionally to the column length and a significant peak broadening will be observed as well because of increased longitudinal diffusion inside the column. One has to keep in mind that the gas molecules are not only traveling in one direction but also sideways and backwards. This broadening is inversely proportional to the flow rate. Broadening is also observed because of the finite rate of mass transfer between the phases and because the molecules are taking different paths through the column.
6. Amount of material injected
Ideally, the peaks in the chromatogram display a symmetric shape (Gaussian curve). If too much of the sample is injected, the peaks show a significant tailing, which causes a poorer separation. Most detectors are relatively sensitive and do not need a lot of material in order to produce a detectable signal. Strictly speaking, under standard conditions only 1-2 % of the compound injected into the injection port passes through the column because most GC instruments are operated in split-mode to prevent overloading of the column and the detector. The splitless mode will only be used if the sample is extremely low in concentration in terms of the analyte.
High temperatures and high flow rates decrease the retention time, but also deteriorate the quality of the separation.
Which detectors are used?
1. Mass Spectrometer (GC/MS)
Many GC instruments are coupled with a mass spectrometer, which is a very good combination. The GC separates the compounds from each other, while the mass spectrometer helps to identify them based on their fragmentation pattern (see Mass Spectrometry chapter).
2. Flame Ionization Detector (FID)
This detector is very sensitive towards organic molecules (10-12 g/s = 1 pg/s, linear range: 106-107), but relative insensitive for a few small molecules i.e., N2, NOx, H2S, CO, CO2, H2O. If proper amounts of hydrogen/air are mixed, the combustion does not afford any or very few ions resulting in a low background signal. If other carbon containing components, are introduced to this stream, cations will be produced in the effluent stream. The more carbon atoms are in the molecule, the more fragments are formed and the more sensitive the detector is for this compound. Unfortunately, there is no direct relationship between the number of carbon atoms and the size of the signal. As a result, the individual response factors for each compound have to be experimentally determined for each instrument. Due to the fact that the sample is burnt (pyrolysis), this technique is not suitable for preparative GC. In addition, several gases are usually required to operate a FID: hydrogen, oxygen (or compressed air), and a carrier gas.
3. Thermal Conductivity Detector (TCD)
This detector is less sensitive than the FID (10-5-10-6 g/s, linear range: 103-104), but is well suited for preparative applications, because the sample is not destroyed. The detection is based on the comparison of two gas streams, one containing only the carrier gas, the other one containing the carrier gas and the compound. Naturally, a carrier gas with a high thermal conductivity i.e., helium or hydrogen is used in order to maximize the temperature difference (and therefore the difference in resistance) between two filaments (= thin tungsten wires). The large surface-to-mass ratio permits a fast equilibration to a steady state. The temperature difference between the reference and the sample cell filaments is monitored by a Wheatstone bridge circuit (the student learnt about this circuitry in physics!).
4. Electron Capture Detector (ECD)
This detector consists of a cavity that contains two electrodes and a radiation source that emits -radiation (i.e., 63Ni, 3H). The collision between electrons and the carrier gas (methane plus an inert gas) produces a plasma-containing electrons and positive ions. If a compound is present that contains electronegative atoms, those electrons will be “captured” to form negative ions and the rate of electron collection will decrease. The detector is extremely selective for compounds with atoms of high electron affinity (10-14 g/s), but has a relatively small linear range (~102-103). This detector is frequently used in the analysis of chlorinated compounds i.e., pesticides (herbicides, insecticides), polychlorinated biphenyls, etc. for which it exhibits a very high sensitivity.