The CHROMacademy Essential Guide Understanding 2D Gas Chromatography
The Essential Guide from LCGC’s CHROMacademy presents an educational webcast on Comprehensive Two Dimensional Gas Chromatography. In this session, Dr. John Hinshaw (CHROMacademy GC Department Dean, GC Column Connections Editor) and Tony Taylor (Technical Director, Crawford Scientific), present a definitive guide to the theory, instrumentation and applications of this now fully established technqiue. The session will consider the need for multidimensional separations in gas chromatography and the application areas in which this technique has found great utility, as well as the various means by which 2D serial column C can be achieved. We will consider the instrumentation required for the comprehensive 2D technique, with special emphasis on column stationary phase and dimensions as well as the design and role of the ‘modulator’ within the system. Detection and data treatment are also carefully considered as these are often the ‘driving force’ behind the successful implementation of the technique form both a qualitative and quantitative standpoint.
A must see for everyone undertaking analysis of complex samples by gas chromatography.
CHROMacademy GC Department Dean
GC Column Connections Editor
Why are multidimensional separations required?
Peak capacity in multidimensional GC
Typical Application Areas
Comprehensive v’s other multidimensional serial column techniques
Traps and modulators – design and important practical aspects
Stationary phase and column dimension choices
MS & other detectors for multidimensional GC
Data reduction / quantitative analysis
Who Should Attend:
Anyone analyzing complex samples by gas chromatography
Key Learning Objectives:
Understand why multidimensional gas chromatography might be required and the advantages it can bring
Appreciate the application areas in which comprehensive 2D gas chromatography is used
Theory and peak capacity considerations of 2D gas chromatography
Describe the various means by which serially coupled columns can be used for multidimensional separation
Investigate the various designs and operational requirements for modulators in comprehensive 2D GC
Appreciate the considerations for column phase and dimension selection for each column
Investigate the various detectors used, with special emphasis on MS detection
Invetigate the various ways in which 2D GC data can be reduced and the ways that the data are presented for both qualitative and quantitative analysis
Multidimensional GC is now an established technique for the analysis of complex samples, with high peak capacity and is widely used in application areas such as petrochemistry, metabonomics, food science, etc.
The technique uses GC columns in ‘series’ to achieve a complete separation of complex samples using orthogonal column chemistries which are either impossible or very time consuming using a single dimensional technique (i.e. using only one GC column). Whilst multi-dimensional GC brings many separation benefits, achieving efficient analyte transfer between columns and the complexity of data analysis are potential barriers to more wholesale adoption as a routine analytical technique.
Chromatographers have used multiple GC columns with orthogonal phase chemistry (in the past a much larger variety of GC stationary phases and supports were available in the ‘packed column’ format) to achieve difficult separations since the late 1960’s. There are two fields of multidimensional GC:
Share this tutorial
Early approaches to multidimensional separations involved the collection of co-eluting compounds prior to a second separation (injection), on a column packed with a different (orthogonal) stationary phase. This earlier form of multidimensional GC, known as off-line multidimensional GC, is no longer widely used as manipulating / manually handling the sample after the first separation can introduce contamination, cause loss of volatile components, discrimination during the second injection and of course will incur further operator interaction time.
Online multidimensional GC, on the other hand, provides the means for a more highly automated separation, removing the need for effluent collection and handling. This approach improves the quality of the result and reduces analysis time. Multidimensional GC instrumentation is technically more complex which is somewhat of a drawback; however, with the advent more advanced technologies, the technique is becoming more accessible to the general laboratory analyst.
The online two dimensional form of the technique (GC×GC) is perhaps the most popular form of multidimensional GC. It utilizes two conventional (usually capillary) GC columns placed in series, either in the same or different ovens. The use of comprehensive two dimensional GC for analyzing samples is not new, with references dating back to the early 1990’s.
Multidimensional separations can be defined as any analytical method where multiple separation mechanisms are employed sequentially.
One Dimensional Separations
In order to illustrate the difference between one and multi-dimensional separations, let us consider a heterogeneous, where each analyte is represented by a geometric figure of certain color.
For illustrative purposes let us assume that we have two completely different chromatographic columns: one capable of separating analytes by size, the other capable of separating analytes by shape.
Figure 3 reveals that three peaks are obtained after separating by shape. Note that each peak contains several analytes (shapes) of different color.
Figure 2. » One dimension separation (by shape)
A similar separation occurs when the second column is used and a separation by size is attempted. However, this time, four different peaks are obtained.
Separation by shape
Figure 3. » One dimensional separation (by size)
Note that neither of the two separations are capable of separating all analytes within the sample.
Separation by shape
“Multidimensional separation techniques are those in which different separation steps or stages, based on different mechanisms, are linked according to certain criteria. The number of different stages can be defined as the dimensionality”.
Let’s take our previous example to the next level by considering that after separating the species by shape (first dimension), we submit only the last peak to separation in the second column. This approach, also known as “heart-cutting”, reveals that our last peak is effectively composed of four analytes rather than one.
Figure 4. » Two dimensional separation –after separation by shape, the third peak is subjected to a second separation by size.
Separation by shape
In a similar way, many heart-cuts can be performed to gain further information regarding the sample:
Figure 5. » Two dimensional separation –multiple “heart-cutting” approach (size and shape) can provide in-depth information about complex samples
Separation by shape
In our example, only three peaks are obtained after separation according to shape. Such simple samples are unlikely to require multidimensional analysis and a suitable thermal gradient with a different stationary phase is likely to be successful; however, this is not always the case and sometimes it is advisable to further investigate everything eluting from the first separation (dimension). This approach is known as comprehensive analysis and produces huge amounts of data, and if this data is handled correctly, detailed information regarding the sample.
Figure 6. Two dimensional separation – comprehensive
two dimensional approach (size and shape)
Figure 7. Three dimensional (GC×GC) chromatogram (Reproduced with permission of LCGC Magazine, LC.GC Europe February 2004). 
In theory, if each dimension is totally orthogonal, then the maximum peak capacity (Φ) can be calculated as the product of the individual peak capacities for each dimension.
Let us consider a situation where the first dimension has a peak capacity of 1000 and the second dimension of 30; the two dimensional GC×GC system would offer a peak capacity of 1000×30 = 30000. To achieve such peak capacity with a one dimension separation, then a 2km GC column would be required (analysis time in the order of 1.5 years).
The amount and complexity of data produced in multidimensional GC makes the analysis of results difficult and time consuming. This situation has motivated the development and use of computer assisted and automated data processing. The results of a 3D model of results from a multidimensional experiment are shown in Figure 7.
As with traditional GC, the following injection techniques are commonly used in multidimensional GC:
Programmed temperature vaporization injection
The first two injection types (split and splitless) dominate the multidimensional GC injection universe.
Sampling techniques such as headspace, solid phase microextraction and pressurized liquid extraction, can be used with multidimensional GC. These techniques are used before sample injection. Whatever technique is used, it is important that the sample is transferred to the column in-tact and in as narrow a band as possible to ensure that losses in efficiency confound neither the first or second dimension separation.
Second dimension columns must achieve separation much faster than their first dimension counterparts in order to optimise the 'sampling rate' from the first dimension and therefore, they tend to be short.
The length of the first column might typically be 20-30 m, the inner diameter 0.25 mm and the film thickness 0.25 μm (what some chromatographers may consider a very ‘standard’ capillary GC column). The second column is typically shorter (1 - 2 m), the inner diameter is narrower (0.1 mm) and the stationary phase is thinner (0.1 μm), to allow faster separations. The reduction in internal diameter is used to counterbalance the decreases in efficiency (plate numbers) obtained from shorter columns.
Stationary phases used in multidimensional GC can be classified in the following way:
Non-polar: include 100% dimethyl polysiloxane and low phenyl (mainly 5% phenyl) columns
Mid-Polar: include high phenyl (35-50% phenyl) and low cyano (6-14% cyano) columns
Polar: include high cyano (more than 50% cyano), trifluoropropyl and polyethylene glycol columns
Other: include columns with chiral stationary phases
It is a common practice to select a non-polar column for the first dimension. Mid-polar columns are very good options for the second dimension; however, polar columns can also be used.
Modulators are used to collect and transfer fractions from the first to the second column. The modulation period (transfer interval) is relatively short (usually 3 – 6 seconds). Modulators are of paramount importance to multidimensional GC and are regarded as the heart of the GC×GC system. There are currently two types of modulators:
Thermal modulators which are further divided into heated and cryogenic types
Thermal and valve modulators are commonly used in multidimensional GC; however, thermal modulators have the additional advantage of concentrating fractions of sample (coming from the first column) before infusion into the second column. This advantage, of course, increases chromatographic efficiency.
The multidimensional GC detection system is used to monitor the passage of the components as they emerge from the column.
As with any chromatographic technique, the detector measures some physico-chemical property of the mobile analyte as it elutes from the column. The response of the detector will change due to changes in the column’s effluent. Traditional GC detectors can be used with multidimensional GC. The most common detection types currently used for multidimensional GC are as follows:
Flame Ionization Detector
Electron Capture Detector
Mass Spectral Detector
As the second dimension column tends to produce very narrow peaks (from a couple of tens to hundreds of milliseconds), detectors with fast data acquisition rates are required (usually from 50 to 200 Hz). Fast second dimension separations are required in order to minimize the modulation time from the first dimension to ensure no sample information is missed. This will be further explained in a later part of this tutorial.
For more information on GC detectors please visit the “GC Detectors” and “Mass analysers” modules from CHROMacademy.
LINK TO GC DETECTORS
LINK TO MASS ANALYSERS
Data Analysis Systems
Due to the huge amount of information typically produced during multidimensional GC separations, specialized hardware and software are normally required to analyze and visualize the data.
This situation is compounded when multidimensional chromatography is coupled to MS, as very large data sets are generated. Manual interpretation of this information is possible, although, it is very time consuming. Several software packages are currently available which can make this task more straightforward.
Most GC×GC system manufacturers provide software solutions to assist with data analysis.
Selected Commercially Available Systems
Selected commercially available GC×GC systems are listed below. For more information, please consult the manufacturer.
Table 1. Selected GC×GC system solutions
Agilent Model 6890 and 7890
Multidimensional/Heartcutting Systems - Single and dual oven
Microfluidic/Dean Switching - three different switching systems available
Cryofocusing - high performance using low-cost liquid CO2 with no chromatographic degradation
Solid state proprietary dual jet CO2 modulator
Automatic column characterization
Dedicated GCxGC HyperChrom software
Complete capillary injector range and injection modes
Fast detector's electronics
Comprehensive GCxGC System
GCxGC chromatography employs two orthogonal GC columns connected in series via a modulator
Effluent from the first column is trapped in the modulator (thermal) for a given period (modulation time) before being focused and released into the second column
No moving parts
Its LN2 cooled thermal modulator supports GC x GC analysis from C3 to C47
PEGASUS 4D ® GCxGC TOFMS
Spectral collection rates up to 500 full-range mass spectra/second (500 Hz)
Cryo-focusing prior to release on the secondary column
Multidimensional GC uses more than one separation stage and is typically performed by coupling two or more capillary GC chromatographic columns (of different stationary phase selectivity) in series.[8, 9, 10]
Elements of GC×GC Systems
Multidimensional GC utilizes two conventional capillary GC columns placed in series. Both columns can be either placed in the same or in different ovens. Modern instruments use either thermal or valve- (time of pressure) based modulators to perform the coupling between the columns. This will be further explained later in the text.
« Figure 8. Multiple-oven two dimensional
(GC×GC) system (oversimplified)
The option to have the second dimension column heated independently of the first brings the possibility of even finer selectivity control however this adds greatly to method development and the complexity of analysis design. Further, as the separation in the second dimension needs to be very rapid, the oven used to heat the second column needs to be very fast, if gradient temperature programming is performed.
It is important to note that second dimension columns must perform their separations much faster than their first dimension counterparts and therefore, they tend to be much shorter. Such short columns will yield fewer theoretical plates (lower efficiency, N), but this is generally not a problem as their function is to separate smaller-numbers of analytes at a time –effectively the unresolved components from each peak in the first dimension separation. The faster one can achieve the second dimension separation the greater the number of ‘cuts’ from the first chromatogram that can be taken (less information wasted) when using the ‘Heart Cutting’ method or the less material that needs to be ‘accumulated’ in the modulator for subsequent resolution in the second column.
Multidimensional GC separations cannot typically be accomplished with fast first dimension columns as, no second dimension column could accomplish the speed requirements for the separation in the second dimension.
One-Dimensional GC Chromatography.
In single-dimensional GC, the sample is carried in a narrow band into the capillary column where the separation takes place based on relative analyte volatility and differences in analyte partition co-efficient (k) between the mobile and stationary phases. The column’s effluent is fed directly into a detector.
The following animation illustrates the principle of standard GC chromatography. On an instrument which is capable of carrying out multidimensional separations.
« Animation 1. 1D GC performed on an instrument capable of 2D operation. This mode is often used to ‘scan’ unknown samples to assess their complexity and the requirements for 2D work.
Multidimensional GC systems are capable of performing one dimensional separations by diverting the effluent from the first column into an 'intermediate' detector.
Such experiments can be used to ‘screen’ samples for their complexity. If the resulting separation of the unknown results in a highly complex chromatogram two options are available:
1. Reduce the initial oven temperature and employ a long, shallow temperature gradient using a stationary phase which is suited to the sample type. Finding a column with suitable selectivity may be very difficult, especially in instances where the volatility and chemistry of the analytes are either very similar or widely different. The use of long shallow gradients (remember that the optimum gradient temperature programming rate in capillary GC is approximated as 10 x t0 / min) obviously leads to very long analysis times.
2. Employ multi-dimensional GC techniques in which the selectivity of the experiment may be ‘tuned’ by an orthogonal phase chemistry for the two columns (typically non-polar in the first dimension and more polar in the second), to cope with the demands of widely differing analyte chemistry. The need for such long shallow temperate gradients is somewhat mitigated as the eluting fractions from the first dimension separation can be quickly and effectively separated in the second dimension using a shorter, (faster) capillary in the second dimension.
Multidimensional GC Chromatography.
In standard multidimensional chromatography, the injected sample flows through the first column where separation takes place according to the ability of the stationary phase to discriminate on sample components volatility and/or chemistry (polarity, ability to hydrogen bond and the degree of π- π electron interaction possible). Typically the first dimension separation will be based on analyte volatility differences. The first column’s effluent is fed into the second column where the second separation takes place. Note that in our ‘generic’ system shown here, each column has an independent detector. A multitude of columns in series can be used instead of just two; however, the complexity of the instrumentation and the data produced precludes more than two columns in all but the most advanced research laboratories.
Note that in comprehensive multidimensional chromatography the whole sample that is injected is separated by the two columns. The following animation illustrates the principle of standard multidimensional chromatography.
« Animation 2. The comprehensive multi-dimensional GC process (oversimplified).
Note that each peak in the chromatogram from detector 1 may result in multiple peaks in detector 2. The data from both detectors can be correlated and plotted to form the "D GC 'map'
A common variation of standard multidimensional chromatography transfers only a fraction of the first column’s output to the second dimension column. This approach, known as “heart cutting”, can be achieved by transferring either single or multiple time based fractions from the first to the second dimension columns. If all fractions exiting the first column are used, then the approach is known as ‘comprehensive multidimensional chromatography’ (see the later sections).
As can be seen from Figure 9, with time based Heart Cutting processes, one relies on the peak(s) of interest eluting within the Heart Cut time – if this does not occur (bottom chromatogram) then potentially crucial information is lost. This is more likely when the peaks resulting from the first dimension separation are narrow (efficient). When broad first dimension peaks are produced, the heart cuts may be taken from some regions across the peak, whereas others may be lost, which obviously risks co-eluting species being lost and not transferred to the second dimension.
Heart cutting is typically used when a modulator (accumulator) device is not available or where the second dimension separation occurs in a time frame which is considerably longer than the accumulation time possible with the modulator device (see later text).
« Figure 9. Loss of information during the heart-cutting approach.
« Animation 3. The heart-cutting process (oversimplified). In this animation, only a fraction of the effluent from the first column is fed into the second column.
In the animation only a single heart-cut was performed so that the first peak is analyzed with both detectors. In this example the second dimension column resolves two components which were not separated using the less polar first dimension phase.
Modulators are used to transfer effluent from the first dimension column to the head of the second dimension column in short repetitive pulses. There are currently two types of modulators: thermal and valve modulators. [6, 10]
« Figure 10. The GCxGC modulation process
Thermal modulators are commonly used in multidimensional GC and have the advantage of being capable of concentrating analyte bands (from the first dimension) prior to infusion into the second. By so doing, the chromatographic efficiency is increased as samples are injected as narrow bands into the second column. The modulation period (transfer interval) is relatively short (usually 2 – 6 seconds). There are currently two types of thermal modulators: heated and cryogenic.
The first modulator designs were based on metal-clad, thick-film capillary columns (modulation tube) that retained and concentrated the analytes eluting from the first column. Once concentrated, analytes were desorbed by resistive heating.
« Figure 11.
Thermal modulator of Liu and Phillips
Lack of robustness in the resistive heating approach led to the development of a rotating heating element (known as a ‘sweeper’) which effectively moved ‘along’ the modulation tube to progress analytes successively through the modulation (trap) tube and into the second dimension column.
« Figure 12. Rotating thermal modulator
The temperature limit of heated modulators prevents the analysis of compounds heavier than around C25 . More modern instruments are based on either valve or cryogenic modulators.
Cryogenic modulators work on a similar way to their heated counterparts. These devices use low temperatures to retain and concentrate the analytes eluting from the first column. The transfer of analytes between the two columns is a two step process:
Focusing step: is usually achieved by a cooling device that retains sample components
Desorption step: the focused sample is then heated to volatilise the analyte components into column 2
Thermal modulators use a cryogenic liquid (such as nitrogen or carbon dioxide) to produce local cold spots where analytes coming from the first column will condense and adsorb. Once focused and accumulated for a fixed length of time, an increase in temperature will cause analyte volatilization and ‘injection’ into the second column.
« Figure 13
GC x GC loop modulator from Shimadzu.
« Animation 4 illustrates the operating principle of a single effect thermal modulator. Focused analytes are infused into the second column as described above, however, any species that flows through the modulator during the desorption (heating) step also flows through to the second column without focusing. This can result in issues with sample co-elution caused by ineffective trapping and releasing (modulation) of analytes which are separated by a time corresponding to less than the duty cycle (heating / cooling cycle) of the modulator. In other words, poorly resolved peaks which elute during the desorption step will have poor efficiency / resolution assing into the second dimension column. This problem can be solved by using dual effect thermal modulators.
The working principle of single effect thermal modulators is explained.
Single action thermal modulators can fail to inject into the second column:
narrow fractions of analytes
only pre-focused analytes
Table 2. Modulation problems
Correct modulation period and transfer between columns
If the modulation period is not carefully controlled, then an increased amount of effluent (analyte) can be transferred between the columns potentially producing ’ghost’ peaks or poor peak resolution.
Peak broadening and poor separation
If the modulator fails to inject narrow bands of analytes into the second column, then peak broadening and poor separation may result.
If the modulation period is not carefully controlled, then a reduced amount of effluent (analyte) can be transferred between the columns, potentially missing vital information.
« Animation 5 Illustrates the operating principle of multiple effect thermal modulators. The alternating effect of the focusing chambers guarantee that only analytes coming from the desorption step are fed into the second column.
Note that the first dimension separation may, under certain circumstances, be adversely affected by the focusing effect on the trapping device (i.e. first dimension separation appears to be ‘lost’ if previously separated analytes are accumulated into the modulator together). However, this is usually a temporary effect, which is overcome by the second generation separation.
The transfer of effluent between columns can also be achieved using one or more valves. The use of valve modulators:
does not involve concentrating the primary effluent
does not require strict temperature control and manipulation as with thermal modulators
is usually simpler, less costly, smaller, and sometimes more rugged than the use of thermal modulators
permits high resolution but may involve a loss of analyte and/or non-optimal flow
« Figure 14
GC and GCxGC with a valve modulator
The diagram below illustrates either traditional (one dimensional) GC or heart-cutting analysis with a commercial valve-based GC x GC instrument. The instrument shown below is based on valves to regulate the transit of eluent and sample within the instrument; however, designs based on thermal regulation are also possible.
« Figure 15. GC×GC system representation (oversimplified)
The number of heart-cuts from the first column that can be sent to the second column is limited by the analysis and recovery times required by the second column. Analytes that are eluted from the first column while the second column is still analyzing a previous heart-cut are sent to the first detector and therefore, they don’t undertake any further chromatographic separation.
There are a few solutions to this problem:
1. Use multiple traps and infuse analytes into the second column only after it finishes its previous heart-cut cycle.
« Figure 16.
GC×GC system with multiple traps
2. Using more than one secondary column (parallel arrangement)
« Figure 17.
GC×GC system with multiple secondary columns in parallel arrangement
3. Speed up the second separation in such a way that peaks do not accumulate (usually shortening the second column)
« Figure 18.
GC×GC heart-cutting configuration with a reduced dimension secondary column
Valve modulators vent primary-column effluent to a certain extent; thus only a fraction of the entire sample undergoes separation in both dimensions. Therefore, comprehensive GCxGC cannot be achieved with such modulators. Serious concerns have been raised regarding valve-based GCxGC for quantitative analysis. However, the amount of information lost in transfer between columns can be minimized by sampling the primary effluent more frequently i.e. ensuring the modulation period is not too large (usually in the order of 1 Hz).
Diaphragm valves are currently the only multiport valves capable of achieving continuous switching at a rate of 1 Hz. Unfortunately, the structural integrity of these valves can be seriously deteriorated at temperatures exceeding 200oC. This restricts the range of samples that can be analyzed.
The figure below illustrates a widely used diaphragm valve, the four port, two position valve with direct injection.
«Figure 19. Four port, two position valve with direct injection.
«Figure 20. Multiport, two position valve with sample loop.
Valves are easy to use and relatively inexpensive. However, they do present certain issues which might include:
peak dispersion in dead volume
poor resistance to high temperatures
As well as valve modulators, pressure sensitive systems that eliminate the aforementioned problems (and that are also less expensive) are currently available. The most popular is the ‘Deans switch’ device.
Deans Switch and Other Modulation Approaches
Invented by D. R. Deans in 1981, the Deans switch is a valve-less system; using three T-pieces and a pneumatic switch between the columns. In essence, the flow direction is controlled by pressure changes on the different T-pieces. Whilst these devices are difficult to set-up in terms of balancing pressure and flow, most manufacturers now use electronic pressure control devices with simple software interfaces to control the device.
Deans switch, survey run (sample is analyzed with first detector only)
Deans switch, sample transfer (if required, sample can be analyzed with both detectors)
Deans switch, first column back-flush
The Deans switch works base on pressure differences to achieve flow diversion. The interactive diagram below, shows how this is achieved.
« Figure 24. Deans switch operating modes.
There are simpler pressure based modulator designs, a typical example is shown in Figures 24 to 27.
« Figure 25.
Simple three T-pieces modulator design
« Figure 26.
Simple three T-pieces modulator design. Fill mode
The solenoid valve permits the correct selection of gas pressure to be supplied (P1 relative to P2) in such a way that the flow from the primary column is restricted and flow at a certain speed while filling the loop.
« Figure 27.
Simple three T-pieces modulator design. Flushing mode
After a certain period of time (sample loop filled), the solenoid valve is acted so the sample loop content is diverted into the second column.
The use of an MS detector will generate a considerable amount of additional information. In fact, this situation can be regarded as having an additional dimension.
The MS detector is maintained under vacuum. Compounds eluting from the GC column are bombarded with electrons (EI) or reagent gas ions (CI) in order to ionize them. Compounds fragment into characteristic charged ions or fragments and the resulting charged species are focused and accelerated into a mass analyzer. Due to its high sampling rate range, Time-of-Flight (TOF) mass analyzers are becoming the standard for multidimensional GCxGC-MS.
Advantages of time of flight mass analyzers include:
Scanning Speed: High sampling rate (50 – 500 Hz)
Mass Range: Highest mass range of all MS analysers
Detection Limit: Low detection limit (usually in the order of pico-grams)
Performance: Increased mass accuracy and resolution
Figure 28 » The use of MS as a detector will add another dimension to the GCxGC data
Disadvantages of time of flight mass analyzers include:
Price: While the price of a conventional quadrupole mass analyser is in the order of 100.000 American dollars, a conventional TOF system is in the order of 250.000 American dollars
Performance: Classical TOF instruments have lower resolution
« Animation 6.
Elements and working principle of time of flight mass analyzers.
« Figure 29. Due to its fast acquisition rates TOF mass analyzers are better suited for GCxGC applications than quadrupole or other classical 'scanning' mass analysers
The GCxGC-MS coupling between columns is performed in the same way already described. The figure below shows a successful dual valve system that permits the coupling of columns with very different eluent flow rate requirements.
The most common representation for GC×GC data is a two dimensional plot, where each axis represents the separation on one column. Contour plots using elevation lines or color coding represent signal intensity, which can also be used for further sample elucidation.
Figure 31. Two dimensional (GC×GC) chromatogram.
Figure 32. Three dimensional (GC×GC) chromatograms. 
Three dimensional plots are also common. Here, the x and y axis represent the separation on each column while the intensity of each peak is reflected on the z axis.
In the diagram, the peak obtained after the first separation, actually contains three different analytes. The use of a second dimension column of different stationary phase selectivity will help to separate those analytes. The amount of analyte transferred (and hence the amount of data collected), will determine the quality of the information gathered regarding the nature of the sample. If only one heart-cut was performed at t=t1, then only one analyte would be transferred to column two and subsequently detected (instead of three). However, if several heart-cuts are performed (or indeed if ALL material is transferred to column two and all data is collected), then a comprehensive amount of information regarding the sample is collected.
Figure 33 » Two dimensional GC×GC modulation process.
The data system must be capable of collecting information rapidly from the detector and rendering it in a form which clearly shows the nature of the separation in two dimensions as shown in a simplified form in Figure 34.
The use of data systems to clearly present the information gained from this technique has been, and to some extent continues to be, one of the major hurdles to overcome. As the technique continues to mature, more intuitive data systems with increased computing power and increasingly sophisticated ways of visualizing data will help to widen its applicability.
Figure 34 » Data transformation and visualization in GC×GC using contour plots in the bottom figure to further clarify the data.[5, 15]
Compound identification is of paramount importance to multidimensional GC. As with any other chromatographic technique, no reliable quantitative data can be obtained without reliable identification. The strategies for analyte identification in multidimensional GC do not differ greatly from those in traditional GC and nclude:
Retention time comparison with standards (when available)
Standard addition (spiking)
MS detection (usually TOF mass analysers)
Comparison of detection profiles (for example FID vs. MS)
When multidimensional chromatography is coupled to MS, large data sets are generated. Manual interpretation of this information can be performed, although, it is very time consuming. Several software packages that facilitate this task are available; some of them are listed below.
Table 3. Selected GC×GC software solutions
Zoex, Houston, TX, USA
Fortner Software, Sterling, VA, USA
Thermo Scientific Waltham, MA, USA
Leco, St. Joseph, MI, USA
Before attempting quantitative analysis in multidimensional GC (as with any chromatographic technique), the relationship between signal response and the analyte concentration must be established.
In traditional GC, quantitative analysis is usually straightforward, as detector response is based on a single chemical species and either peak area or peak height can be used. The situation is different in multidimensional GC, where not only peak area and peak height but peak volume can be used as an indication of the amount of analyte responsible for the signal.
Peak area of the modulated compound is a sum of the individual peaks generated by the modulation process
Peak height can be found from the individual peaks in the chromatogram
Peak height is rarely determined from a 3D plot
Peak volume can be determined from a 3D plot and is commonly used
In addition to single analytes, multidimensional GC is well suited for the identification of groups of compounds. The technique is commonly used in petrochemical analysis. For example in order to identify and characterize the various analyte types present as well as a means for easily identifying anomalies or outliers within the sample.
« Figure 35 . GC×GC–FID chromatograms of a diesel on a polar (first dimension) × medium-polar (second dimension) columns.
Of overriding importance to quantitative 2D GC is the effect of matrix components on the quantification (peak overlapping) and secondary column overloading.
Sample fractions eluting from the first column can overload the second column if excessive amounts of solute are infused.
In multidimensional GC, overloading can be identified by asymmetric two dimensional signals.
« Figure 36. Normal versus overloaded peaks in multidimensional GC.
Mass transfer between the columns is another issue to consider. Due to the focusing effect (within the modulator), and a rapid, efficient separation in the second dimension, peak height in multidimensional GC can be greater than in traditional GC, but peak areas should be comparable.
Peak area from 2D GC signals can be used to form calibration curves, providing the peaks can be accurately identified, which is possible using standard solutions just as in the 1D technique
Multidimensional GC is becoming important in the analysis of samples with large numbers of components especially where the samples contain groups of analytes with similar volatility or polarity. In fact, it was the analysis of crude oil fractions that was at the forefront of multidimensional GC development.
To list a full range of multidimensional GC application areas is prohibitive since its flexibility makes it suitable to a multitude of application types. Examples of interesting applications are shown below:
Petrochemistry, fuels and power generation
Flavors and fragrances
Environmental analysis and pollutants
Petrochemistry, Fuels and Power Generation
The GCxGC analysis of a diesel sample is shown in Figure 37. Six different series of compounds are highlighted (phenanthrenes, tetrahydro-aceanthrecenes, pyrenes, tetrahydro-chrysenes and chrysenes), each of them containing a large number of analytes. GCxGC allows the identification of not only individual compounds but also groups of species. In this way potential contaminants or outliers are more easily identified. This is an ideal way to characterize samples and is invaluable for ‘reverse engineering’ samples.
« Figure 37. GCxGC analysis of a diesel sample.
Columns: Primary weak-polar column SPB-5 (30m×0.25mm×1.0μm), secondary mid-polar column BPX-50 (3m × 0.1mm × 0.25μm). Oven temperature program: 60 ◦C for 0 min, 3 ◦C/min to 300 ◦C (0 min hold); helium flow: 1.2 mL/min Injection mode: split 75:1; 45 psi at 1.1 min; injection temperature: 60 ◦C; injection volume: 0.2μL. Modulation period: 10 s Detection: TPF-MS, sampling rate for the detector 100Hz
The GCxGC analysis of an apple extract is shown in Figure 38. In this particular application, twenty related pesticides are successfully separated and identified in less than 2 minutes. This fast visual screen can be used to identify samples containing target compounds which are then submitted for a more comprehensive analysis and quantitation.
« Figure 38. GC×GC analysis of pesticides from an apple extract.
Columns: Primary column DB-XLB (30m×0.25mm×0.25μm), secondary column DB-17 (1m × 0.1mm × 0.1μm). Oven temperature program: 70 ◦C for 1.1 min, 20 ◦C/min to 300 ◦C, 20–300 ◦C for 10 min; helium flow: 1.2 mL/min Injection mode: pulsed splitless; 50 psi at 1.1 min; injection temperature: 250 ◦C; injection volume: 1μL. Detection: FID, sampling rate for the detector 300Hz
The GCxGC analysis of a volatile oil, rich in compounds of potential pharmaceutical interest, is shown in Figure 39. The total analysis time is roughly 80 mins (4750 seconds); however, over 800 compounds were successfully analyzed.
« Figure 39. GC×GC analysis of Pogostemon cablin Benth (volatile oil used in traditional Chinese medicines).
Columns: Primary column PEG (60m×0.25mm×0.25μm), secondary column Cyclodextrine (3m × 0.1mm × 0.1μm). Oven temperature program: 70 ◦C for 3.0 min, 3 ◦C/min to 200 ◦C, hold for 35 min; helium flow: 1.2 mL/min Injection mode: Not provided. Detection: FID, sampling rate for the detector 100Hz
Flavors and Fragrances
The GCxGC analysis of a grape extract is shown in Figure 40. The total analysis time is roughly 50 mins; however, 60 related compounds (terpenols, terpendiols, esters and aldehydes) were successfully separated. Applications of this type are invaluable in metabonomic and metabolomics work, where metabolites of interest are more easily indentified and tracked than with 1D techniques.
« Figure 40.GC×GC extracted ion chromatogram contour plot of m/z 93, 121 and 136 from a grape extract sample.
Columns: Primary non-polar column (60m×0.25mm×0.25μm), secondary column polar (2.5m × 0.1mm × 0.1μm). Injection mode: Splitless 250◦C. Modulation period: 4 s Primary oven temperature program: 40 ◦C for 1.0 min, 5 ◦C/min to 260 ◦C (15 min hold); helium flow: 1.7 mL/min Secondary oven temperature program: 45 ◦C for 1.0 min, 5 ◦C/min to 265 ◦C (15 min hold); helium flow: 1.7 mL/min Detection: TOF-MS, sampling rate for the detector 50Hz
Environmental Analysis and Pollutants
The GCxGC analysis of polycyclic aromatic hydrocarbons (PAH’s) from a diesel sample is shown in Figure 41. The total analysis time is roughly 93 mins; however, 19 closely related compounds (see the sample details below) were successfully analyzed.
Figure 41. GCxGC analysis of a polycyclic aromatic hydrocarbons (PAH’s), sample.
Zaiyou Liu and John B. Phillips. “Comprehensive Two-Dimensional Gas Chromatography using an On-Column Thermal Modulator Interface” Journal of Chromatographic Science, Jun-1991, Volume 29, Pp 227-231
J. Calvin Giddings. “Sample Dimensionality: A Predictor Of Order-Disorder in Component Peak Distribution in Multidimensional Separation” Journal of Chromatography A, 703 (1995) 3-15
Stephen E. Reichenbach, Mingtian Ni, Visweswara Kottapalli, Arvind Visvanathan. “Information Technologies for Comprehensive Two-Dimensional Gas Chromatography” Chemometrics and Intelligent Laboratory Systems 71 (2004) 107– 120
Luigi Mondello, Peter Quinto Tranchida, Paola Dugo, and Giovanni Dugo. “Comprehensive Two-Dimensional Gas Chromatography-Mass Spectroscopy: A Review” Mass Spectrometry Reviews, 2008, 27, 101– 124
M. Adahchour, J. Beens, R. J. J. Vreuls, U. A. Th. Brinkman. “Recent Developments in Comprehensive Two-Dimensional Gas Chromatography (GC×GC) I. Introduction and Instrumental Set-Up” Trends in Analytical Chemistry, Pp 438 – 454. Vol. 25, No. 5, 2006
M. Adahchour, J. Beens, R. J. J. Vreuls, U. A. Th. Brinkman. “Recent Developments in Comprehensive Two-Dimensional Gas Chromatography (GC×GC) II. Modulation and Detection” Trends in Analytical Chemistry, Pp 438 – 454. Vol. 25, No. 5, 2006
“GC Detectors” from “Theory and Instrumentation of GC”, GC Channel from CHROMacademy.
John V. Hinshaw. “Comprehensive Two-Dimensional Gas Chromatography” GC Connections. LC-GC Europe February 2004. Pp 2-7
Matthew S. Klee. “Retention-Time Locked Comprehensive Multidimensional Gas Chromatography” US Patent No 6494078B1. Dec 17, 2002.
Liu Z. and Phillips J. B. “Comprehensive Two-Dimensional Gas Chromatography Using an On-Column Thermal Modulator Interface” Journal Chromatography Science. Vol 29. Pp 227-231. 1991.
Ronald D. Snelling, Clifford M. Taylor, Zhuangzhi ‘Max’ Wang, Richard R. Whitney. “New Applications for Comprehensive Two Dimensional Gas Chromatography”. Pp 1-18. Shimadzu 2010
James D. McCurry and John V. Seeley. “Single Stage Flow Modulator For Performing Comprehensive Chromatography”. US Patent No 7383718 B2. June 10 2008.
Marina Poliak, Alexander B. Fialkov, Aviv Amirav. “Pulsed Flow Modulation Two-Dimensional Comprehensive Gas Chromatography–Tandem Mass Spectrometry With Supersonic Molecular Beams” Journal of Chromatography A, 1210 (2008) 108–114
Jens Dallüge, Jan Beens and Udo A.Th. Brinkman. “Comprehensive two-Dimensional Gas Chromatography: a Powerful and Versatile Analytical tool” Journal of Chromatography A, 1000 (2003) 69–108
C. Vendeuvre, R. Ruiz-Guerrero, F. Bertoncini, L. Duval and D. Thiébaut. “Comprehensive Two-Dimensional Gas Chromatography for Detailed Characterisation of Petroleum Products” Oil & Gas Science and Technology. Vol 62 (2007), No 1, Pp 43-55
Frank Cheng-Yu Wang, Kuangnan Qian, and Larry A. Green. “GC×MS of Diesel: A Two-Dimensional Separation Approach” Analytical Chemistry 2005, 77, 2777-2785
Jitka Zrostlíková, Jana Hajšlová b, Tomáš Čajka. “Evaluation of Two-Dimensional Gas Chromatography–Time-Of-Flight Mass Spectrometry for the Determination of Multiple Pesticide Residues in Fruit” Journal of Chromatography A, 1019 (2003) 173–186
Jianfang Wu, Xin Lu, Wanying Tang, Hongwei Kong, Shenfan Zhou, Guowang Xu. “Application of Comprehensive Two-Dimensional Gas Chromatography–Time-Of-Flight Mass Spectrometry in the Analysis of Volatile Oil of Traditional Chinese Medicines” Journal of Chromatography A, 1034 (2004) 199–205
Sílvia M. Rocha, Elisabete Coelho, Jitka Zrostlíková, Ivonne Delgadillo, Manuel A. Coimbr. “Comprehensive Two-Dimensional Gas Chromatography With Time-Of-Flight Mass Spectrometry of Monoterpenoids as a Powerful Tool for Grape Origin Traceability” Journal of Chromatography A, 1161 (2007) 292–299
Daniela Cavagnino, Paolo Magni, Giacinto Zilioli, Sorin Trestianu. “Comprehensive Two-Dimensional Gas Chromatography Using Large Sample Volume Injection for the Determination of Polynuclear Aromatic Hydrocarbons in Complex Matrices” Journal of Chromatography A, 1019 (2003) 211–220
As a member of CHROMacademy, you will also get access to LCGC magazine articles from your favourite authors - John Dolan, John Hinshaw, Mike Balough, and Ron Majors; over 300 news items refreshed daily; vendor application notes, webcasts and podcasts; electronic laboratory tools and calculators; and access to the growing Essential Guides archive, recorded tutorials by industry experts.
Subscribe now for $399
and get instant access to all CHROMacademy Essential Guide Webcasts
The following subjects are covered in CHROMacademy.com
The Theory Of HPLC
Chromatographic Parameters (3hrs)
Band Broadening (3hrs)
Column chemistry (4hrs)
Reverse phase (partition) chromatography (6hrs)
Ion-Pair Chromatography (3hrs)
Normal phase (absorption) chromatography (3hrs)
Gradient HPLC (3hrs)
Quantitative and Qualitative HPLC (3hrs)
FAST HPLC (4.5hrs)
Theory and Instrumentation of GC
Chromatographic Parameters (3hrs)
Band Broadening (3hrs)
Gas Supply and Pressure Control (2hrs)
Sampling Techniques (4.5hrs)
Sample Introduction (5hrs)
GC Columns (5.5hrs)
GC Temperature Programming (3hrs)
GC Detectors (2.5hrs)
Instrumentation of HPLC
Mobile Phase Considerations (3.5hrs)
Solvent Pumping Systems (4hrs)