Gas Chromatography (GC)

Introduction

Gas chromatography is a chromatographic technique that can be used to separate volatile

organic compounds. A gas chromatograph consists of a flowing mobile phase, an
injection port, a separation column containing the stationary phase, and a detector. The
organic compounds are separated due to differences in their partitioning behavior
between the mobile gas phase and the stationary phase in the column.

Instrumentation

Mobile phases are generally inert gases such as helium, argon, or nitrogen. The injection
port consists of a rubber septum through which a syringe needle is inserted to inject the
sample. The injection port is maintained at a higher temperture than the boiling point of
the least volatile component in the sample mixture. Since the partitioning behavior is
dependant on temperture, the separation column is usually contained in a thermostat-
controlled oven. Separating components with a wide range of boiling points is
accomplished by starting at a low oven temperture and increasing the temperture over
time to elute the high-boiling point components. Most columns contain a liquid stationary
phase on a solid support. Separation of low-molecular weight gases is accomplished with
solid adsorbents. Separate documents describe some specific GC Columns and GC
Detectors.

Schematic of a gas chromatograph

Gas Chromatography Columns

Columns

Gas chromatography columns are of two designs: packed or capillary. Packed columns
are typically a glass or stainless steel coil (typically 1-5 m total length and 5 mm inner
diameter) that is filled with the stationary phase. Capillary columns are a thin fused-silica
(purified silicate glass) capillary (typically 10-100 m in length and 250 µm inner
diameter) that has the stationary phase coated on the inner surface. Capillary columns
provide much higher separation efficiency than packed columns but are more easily
overloaded by too much sample.

Gas Chromatography (GC) Detectors

Introduction

After the components of a mixture are separated using gas chromatography, they must be
detected as they exit the GC column. The links listed below provide the details of some
specific GC detectors. The thermal-conductivity (TCD) and flame-ionization (FID)
detectors are the two most common detectors on commercial gas chromatographs. The
requirements of a GC detector depends on the separation application. For example, one
analysis might require a detector that is selective for chlorine-containing molecules,
another analysis might require a detector that is non-destructive so that the analyte can be
recovered for further spectroscopic analysis.

Specific GC detectors

Atomic-emmision detector (AED)

Chemiluminescence detector

Electron-capture detector (ECD)

The ECD is as sensitive as the FID but has a limited dynamic range and finds its
greatest application in analysis organic molecules that contain electronegative
functional groups, such as halogens, phosphorous, and nitro groups.
Flame-ionization detector (FID)
The FID is extremely sensitive with a large dynamic range, its only disadvantage
is that it destroys the sample.
Flame-photometric detector (FPD)

Photoionization detector (PID)

Thermal conductivity detector (TCD)

The TCD is not as sensitive as other dectectors but it is non-specific and non-
destructive.

Atomic-Emission Detector (AED)

Introduction
As capillary column based gas chromatography takes its place as the major, highest
resolution separation technique available for volatile, thermally stable compounds, the
requirements for the sensitive and selective detection of these compounds increases.
Since more and more complex mixtures can be successfully separated, subsequent
chromatograms (output of a chromatographic separation) are increasingly more complex.
Therefore, the need to differentiate between the sample components using the GC
detector as a means of compounds discriminating is more and more common. In addition,
each detector has its own characteristics (selectivity, sensitivity, linear range, stability,
cost, etc.) that helps in a decision about which detector to use.

One of the newest additions to the gas chromatographer's arsenal is the atomic emission
detector (AED). This detector, while quite expensive compared to other commercially
available GC detectors, is an extremely powerful alternative. FOR INSTANCE, Instead
of measuring simple gas phase (carbon containing) ions created in a flame as with the
flame ionization detector, or the change in background current because of electronegative
element capture of thermal electrons as with the electron capture detector, the AED has a
much wider applicability because it is based on the detection of atomic emissions.

The strength of the AED lies in the detector's ability to simultaneously determine the
atomic emissions of many of the elements in analytes that elute from a GC capillary
column (called eluants or solutes in some books). As eluants come off the capillary
column they are fed into a microwave powered plasma (or discharge) cavity where the
compounds are destroyed and their atoms are excited by the energy of the plasma. The
light that is emitted by the excited particles is separated into individual lines via a
photodiode array. The associated computer then sorts out the individual emission lines
and can produce chromatograms made up of peaks from eluants that contain only a
specific element.

Instrumentation

The components of the AED include 1) an interface for the incoming capillary GC
column to the microwave induced plasma chamber, 2) the microwave chamber itself, 3) a
cooling system for that chamber, 4) a diffraction grating and associated optics to focus
then disperse the spectral atomic lines, and 5) a position adjustable photodiode array
interfaced to a computer. The microwave cavity cooling is required because much of the
energy focused into the cavity is converted to heat.
Schematic of a gas chromatographic atomic emission detector

Chemiluminescence Spectroscopy
Introduction

Chemiluminescence, like atomic emission spectroscopy (AES), uses quantitative

measurements of the optical emission from excited chemical species to determine analyte
concentration; however, unlike AES, chemiluminescence is usually emission from
energized molecules instead of simply excited atoms. The bands of light determined by
this technique emanate from molecular emissions and are therefore broader and more
complex then bands originating from atomic spectra. Furthermore, chemiluminescence
can take place in either the solution or gas phase, whereas AES is almost strictly as gas
phase phenomenon.

Though liquid phase chemiluminescence plays a significant role in laboratories using this
analytical technique (often in conjunction with liquid chromatography), we will
concentrate on gas phase chemiluminescence reactions since the instrumental
components are somewhat simpler. These detectors are also often used as detectors for
gas chromatography.

Like fluorescence spectroscopy, chemiluminescence's strength lies in the detection of

electromagnetic radiation produced in a system with very low background. And on top of
this, because the energy necessary to excite the analytes to higher electronic, vibrational,
and rotational states (from which they can decay be emission) does not come from an
external light source like a laser or lamp, the problem of excitation source scattering is
completely avoided. The major limitation to the detection limits achievable by
chemiluminescence involves the dark current of the photomultiplier (PMT) necessary to
detect the analyte light emissions.

If the excitation energy for analytes in chemiluminescence doesn't come from a source
lamp or laser, where does it come from? The energy is produced by a chemical reaction
of the analyte and a reagent. An example of a reaction of this sort is shown below:

A chemiluminescence reaction

In gas phase chemiluminescence, the light emission (represented as Planck's constant

times nu-the light's frequency) is produced by the reaction of an analyte (dimethyl sulfide
in the above example) and a strongly oxidizing reagent gas such as fluorine (in the
example above) or ozone, for instance. The reaction occurs on a time scale such that the
production of light is essentially instantaneous; therefore, most analytical systems simply
mix analytes and the reagent in a small volume chamber directly in front of a PMT. If the
analytes are eluting from a gas chromatographic column then the end of the column is
often fed directly into the reaction chamber itself. Since as much of the energy released
by the reaction should (in the analyst's eye) be used to excite as many of the analyte
molecules as possible, loss of energy via gas phase collisions is undesirable, and
therefore a final consideration is that the gas pressure in the reaction chamber be
maintained at a low pressure (~ 1 torr) by a vacuum pump in order to minimize the
effects of collisional deactivation.

It must be stated that the ambiguous specification of "products" in the above reaction is
often necessary because of the nature and complexity of the reaction. In some reactions,
the chemiluminescent emitters are relatively well known. In the above reaction the major
emitter is electronically and vibrationally excited HF; however, in the same reaction,
other emitters have been determined whose identities are not known and these also
contribute to the total light detected by the PMT.

To the analytical chemist the ambiguity about the actual products in the reaction is, in
most case, not important. All the analyst cares about is the sensitivity of the instrument
(read detection limits for target analytes), its selectivity-that is, response for an analyte as
compared to an interfering compound, and the linear range of response.

Here is a schematic of the components necessary for a gas phase chemiluminescence

detector interfaced to a capillary gas chromatograph.
Schematic of a GC chemiluminescence detector

Electron Capture Detectors(ECD)

Introduction

The ECD uses a radioactive Beta emitter (electrons) to ionize some of the carrier gas and
produce a current between a biased pair of electrodes. When organic molecules that
contain electronegative functional groups, such as halogens, phosphorous, and nitro
groups pass by the detector, they capture some of the electrons and reduce the current
measured between the electrodes. The ECD is as sensitive as the FID but has a limited
dynamic range and finds its greatest application in analysis of halogenated compounds.

Schematic of an ECD

Flame-Ionization Detectors (FID)

Introduction

An FID consists of a hydrogen/air flame and a collector plate. The effluent from the GC
column passes through the flame, which breaks down organic molecules and produces
ions. The ions are collected on a biased electrode and produce an electrical signal. The
FID is extremely sensitive with a large dynamic range, its only disadvantage is that it
destroys the sample.

Schematic of FID

Flame Photometric GC Detector

Introduction

The reason to use more than one kind of detector for gas chromatography is to achieve
selective and/or highly sensitive detection of specific compounds encountered in
particular chromatographic analyses. The determination of sulfur or phosphorus
containing compounds is the job of the flame photometric detector (FPD). This device
uses the chemiluminescent reactions of these compounds in a hydrogen/air flame as a
source of analytical information that is relatively specific for substances containing these
two kinds of atoms. The emitting species for sulfur compounds is excited S2. The lambda
max for emission of excited S2 is approximately 394 nm. The emitter for phosphorus
compounds in the flame is excited HPO (lambda max = doublet 510-526 nm). In order to
selectively detect one or the other family of compounds as it elutes from the GC column,
an interference filter is used between the flame and thephotomultiplier tube (PMT) to
isolate the appropriate emission band. The drawback here being that the filter must be
exchanged between chromatographic runs if the other family of compounds is to be
detected.

Instrumentation

In addition to the instrumental requirements for 1) a combustion chamber to house the

flame, 2) gas lines for hydrogen (fuel) and air (oxidant), and 3) an exhaust chimney to
remove combustion products, the final component necessary for this instrument is a
thermal (bandpass) filter to isolate only the visible and UV radiation emitted by the
flame. Without this the large amounts of infrared radiation emitted by the flame's
combustion reaction would heat up the PMT and increase its background signal. The
PMT is also physically insulated from the combustion chamber by using poorly
(thermally) conducting metals to attach the PMT housing, filters, etc.

The physical arrangement of these components is as follows: flame (combustion)

chamber with exhaust, permenant thermal filter (two IR filters in some commercial
designs), a removable phosphorus or sulfur selective filter, and finally the PMT.

Schematic of a gas chromatographic flame photometric detector

Photoionization Detector
Introduction
The reason to use more than one kind of detector for gas chromatography is to achieve
selective and/or highly sensitive detection of specific compounds encountered in
particular chromatographic analyses. The selective determination of aromatic
hydrocarbons or organo-heteroatom species is the job of the photoionization detector
(PID). This device uses ultraviolet light as a means of ionizing an analyte exiting from a
GC column. The ions produced by this process are collected by electrodes. The current
generated is therefore a measure of the analyte concentration.

Theory

If the energy of an incoming photon is high enough (and the molecule is quantum
mechanically "allowed" to absorb the photon) photo-excitation can occur to such an
extent that an electron is completely removed from its molecular orbital, i.e. ionization.

A Photoionization Reaction

If the amount of ionization is reproducible for a given compound, pressure, and light
source then the current collected at the PID's reaction cell electrodes is reproducibly
proportional to the amount of that compound entering the cell. The reason why the
compounds that are routinely analyzed are either aromatic hydrocarbons or heteroatom
containing compounds (like organosulfur or organophosphorus species) is because these
species have ionization potentials (IP) that are within reach of commercially available
UV lamps. The available lamp energies range from 8.3 to 11.7 ev, that is, lambda max
ranging from 150 nm to 106 nm. Although most PIDs have only one lamp, lamps in the
PID are exchanged depending on the compound selectivity required in the analysis.

Selective detection using a PID

Here is an example of selective PID detection: Benzene's boiling point is 80.1 degrees C
and its IP is 9.24 ev. (Check the CRC Handbook 56th ed. page E-74 for IPs of common
molecules.) This compound would respond in a PID with a UV lamp of 9.5 ev
(commercially available) because this energy is higher than benzene's IP (9.24).
Isopropyl alcohol has a similar boiling point (82.5 degrees C) and these two compounds
might elute relatively close together in normal temperature programmed gas
chromatography, especially if a fast temperature ramp were used. However, since
isopropyl alcohol's IP is 10.15 ev this compound would be invisible or show very poor
response in that PID, and therefore the detector would respond to one compound but not
the other.

Instrumentation
Since only a small (but basically unknown) fraction of the analyte molecules are actually
ionized in the PID chamber, this is considered to be a nondestructive GC detector.
Therefore, the exhaust port of the PID can be connected to another detector in series with
the PID. In this way data from two different detectors can be taken simultaneously, and
selective detection of PID responsive compounds augmented by response from, say, an
FID or ECD. The major challenge here is to make the design of the ionization chamber
and the downstream connections to the second detector as low volume as possible (read
small diameter) so that peaks that have been separated by the GC column do not broaden
out before detection.

Schematic of a gas chromatographic photoionization detector

Thermal Conductivity Detectors (TCD)

Introduction

A TCD detector consists of an electrically-heated wire or thermistor. The temperature of

the sensing element depends on the thermal conductivity of the gas flowing around it.
Changes in thermal conductivity, such as when organic molecules displace some of the
carrier gas, cause a temperature rise in the element which is sensed as a change in
resistance. The TCD is not as sensitive as other dectectors but it is non-specific and non-
destructive.

Instrumentation
Two pairs of TCDs are used in gas chromatographs. One pair is placed in the column
effluent to detect the separated components as they leave the column, and another pair is
placed before the injector or in a separate reference column. The resistances of the two
sets of pairs are then arranged in a bridge circuit.

Schematic of a bridge circuit for TCD detection

The bridge circuit allows amplification of resistance changes due to analytes passing over
the sample thermoconductors and does not amplify changes in resistance that both sets of
detectors produce due to flow rate fluctuations, etc.

Gas Chromatographic Injectors

Introduction

The great analytical strength of capillary gas chromatography lies in its high resolution.
Capillary columns have 1) more theoretical plates (a measure of column resolving power
or efficiency) per meter as compared to packed columns and 2) since they have less
resistance to flow they can be longer than packed columns. This means that the average
capillary column (30 meters long) has approximately 100,000 theoretical plates while the
average packed column (3 meters) has only 2500 plates.

But with this separation power comes some limitations: 1) Capillary columns, because
they have smaller diameters (0.05 to 0.53 mm) than packed columns (2 to 4 mm), require
relatively specialized injectors and ancillary flow and pressure controllers and 2)
capillary column require a smaller amount of sample than packed columns. While the
average sample mass of each component in a mixture that is separable by packed column
GC can be in the microgram range (10-6 grams) per injection, capillary columns routinely
only handle 50 nanograms (10-9 grams) of a particular component or less.

Overloaded Chromatography

This sample size requirement initially meant that if samples contained components that
were too concentrated for a capillary chromatographic analysis, the sample had to be
diluted before it was analyzed. Otherwise the column would be overloaded by those high
concentrated components. An example of this appears in the first figure below. The
clearly overload peaks are indicated. And while some of the other components are in the
resolvable (not overloaded) range, having large masses of components can also distort the
peak shape of some of the lower mass components.

An Overloaded Chromatogram

The following figure shows a little better chromatography with fewer overloaded peaks.
The second eluting peak (about 6 minutes) is clearly not overloaded while the group
between 10 and 14 minutes still shows overloading characteristics: long drawn-out tailing
and much less than baseline separation with peaks that elute nearby (the 11 and 12
minute peaks, for instance).
Normal Packed Column Injector

The normal sequence of events in a GC injection is as follows. We will assume in this

explanation that some analytes are dissolved in a (liquid) solvent although much of this
process also holds for gas GC injections too: A small amount of liquid (microliters) is
injected through a silicon rubber septum (using a special microliter syringe) into the hot
(usually 200+ degrees C) GC injector that is lined with an inert glass tube. The injector is
kept hot by a relatively large, metal heater block that is thermostatically controlled. The
sample is immediately vaporized and a pressurized, inert, carrier gas-which is continually
flowing from a gas regulator through the injector and into the GC column-sweeps the
gaseous sample, solvent, analyte and all, onto the column. In the packed column injector,
ALL the vaporized sample enters onto the column. This is how all packed column
injectors work; everything that is injected goes onto the column. One modification of this
is a small ancillary flow of carrier gas that bathes the underside of the injector's septum
so that hot vaporized sample gases can't interact and possibly stick to the septum. This
improves peak shape and reproducibility. This last feature is called the septum purge. The
following figure is a schematic of a packed column injector, sometimes called a direct or
flash injector. The septum purge is not shown here although the carrier gas regulator and
inlet, a septum, and an injection port liner ARE detailed. Last by not least, the packed GC
column itself is connected at the bottom of the injector via metal fittings.
Schematic of packed GC column injector with septum purge

One last subtle point about the configuration of the carrier gas inlet: notice that it enters
the injector at about the middle of the heating block and its gas has to travel along the
outside of the injector before it enters the injection port liner AT THE TOP. This is so the
carrier is preheated before it enter the liner where the sample is vaporized. This helps to
prevent a cold spot at the top of the injector where the carrier gas enters.

Now remember that the size of the capillary column limits the amount of analyte that can
be injected, otherwise, chromatographic overloading occurs. Therefore this packed
column injector design, if used with a capillary column, would require that samples with
high concentrations of analytes be diluted. Unless... what other alternative is there to get
the amount of analytes that are injected onto the column smaller without having to dilute
concentrated samples? The solution is the split/splitless capillary GC injector.

The Split/Splitless Capillary Injector

OK so far so good. But how does the split/splitless injector work? It starts with the same
requirements as the packed column injectors: carrier gas inlet, a septum, septum purge,
injector insert, heater block, and column connection; but the heart of this technological
feat is another set of gas lines out of the injector-another path that the vaporized sample
can take. This is called the split line or vent. The manufacturers of these systems design
them so that the carrier gas flow onto the column is constant-to maintain the
chromatographic requirements of the column and yield reproducible retention times for
analytes. At the same time, the amount of gas that goes out the split vent controls the
amount of sample that enters the column. If the split vent is closed, via a computer
controlled split valve, then all of the sample introduced into the injector goes on the
column. If the split vent is open then most of the vaporized sample is thrown away to
waste via the split vent and only a small portion of the sample is introduced to the
column. The following diagram illustrates a split/splitless injector with the split vent on
so that only a small portion of the sample injected goes on the column.

And finally, a very neat aspect of this is that the amount of gas exiting the split vent can
be varied while keeping the flow onto the column constant. This means that the
AMOUNT of the split (called the split ratio) can be varied. A common split ratio is 50 to
1. That is, for every 50 units of gaseous sample that are thrown away to waste, 1 unit
goes on the column.

The analyst keeps careful control of the split ratio so that results from the
chromatography can still be quantified. Chromatographic peaks that show up as, say 2.5
ng of compound X really represent 2.5 x 50 = 125 ng of analyte X in the original sample.
Also notice that this mass (125 nanograms) would have overloaded the column if all of it
ended up on the capillary column. Voila! A split injection, and no sample dilution
required.