Retention index guide
|MassFinder 4: Manual|
MassFinder's Retention Index Guide explains the concept of retention indices in gas chromatography and its application to identify compounds, particularly in combination with mass spectrometry (GC/MS).
This guide is meant as a concise, system-independent introduction, while the tutorial lesson Analysing your own data gives a detailed step-by-step explanation on how to use the retention index feature of MassFinder with your own data files.
Abstract: (1) Retention indices are relative retention times normalised to closely eluting n-alkanes. Retention indices are system independent and long-term reproducible, even after many years and in different laboratories around the world. (2) Identifying peaks by library searches should not solely focus on mass spectral similarity, but also include retention indices in order to optimise the quality and reliablitity of library hits. Many isomeric compounds like sesquiterpene hydrocarbons can only be identified if taking both mass spectra and retention indices into account.
We assume that you are familiar with modern gas chromatography as it is employed in most analytical laboratories nowadays and can be described as high-resolution capillary gas-liquid chromatography. Typically, polysiloxanes are used as stationary phase, and hydrogen or helium is used as mobile phase.
Gas chromatography is an analytical method to separate and identity compounds. Separation occurs due to different gas/liquid equilibrium constants which in turn depend on the polarity and volatility of the analytes. The gas chromatographic retention time can be used as a property to characterise the compound, because under constant chromatographic conditions the retention time of a compound is reproducible.
Identification based on retention times relies on knowing which compound elutes at a certain retention time, i.e. you compare the observed retention time of your sample with a table of previously recorded retention times. Thus, reference compounds of all possible constituents have to be measured under exactly identical chromatographic conditions. Naturally, different compounds might coincidentally co-elute at the same retention time, which somewhat limits the scope of gas chromatography and gives rise to the need of more enhanced analytical methods like GC/MS.
The major problem of the retention time based approach of identifying compounds is the necessity of maintaining exactly identical chromatographic conditions. A subtle temperature difference of 1 °C, a slightly increased carrier gas pressure, or a few seconds of delay when starting the acquisition may cause retention time deviations larger then the retention time range of several possible constituents.
Further, it is not possible with the required degree of accuracy to compare the retention times of one GC with another system, neither in the same laboratory nor worldwide. In most cases, system maintenance like shortening the column or installing a new column will change the retention times and require all reference retention times to be measured again. Routinely, references and samples are run in the same sequence shortly after each other.
In summary, retention times are valuable information to characterise and identify compounds, but they are poorly reproducible long-term or between different systems and should only be relied on when measuring reference and sample under identical conditions and shortly after each other.
Relative Retention Times
One basic approach to overcome these limitations was to calculate relative retention times, i.e. dividing the retention time of your compound by the retention time of an internal standard. Thus, slight variations of temperature can be compensated (because they have equal effect on both compounds) and relative retention times are to a certain degree even comparable between different systems. However, initial acquisition delays or entirely different temperature programs can not be compensated with this method. Further, the larger the retention time difference between internal standard and target compound, the less accurate is this kind of compensation, and the more disturbances may occur after the first compound eluted, thus affecting only one of both substances, which in turn can not be compensated by this method.
The Concept of Retention Indices
These limitations can mostly be resolved by calculating relative retention times based on two internal standards, one shortly eluting before and the other shortly eluting after the target compound. Thus, quite a large number of standards is necessary to cover the complete time range and the use of inert n-alkanes is established for this purpose. The difference between the retention times of two consecutive n alkanes is divided in 100 parts and the so-called retention index of an n-alkane itself is defined as 100 n.
Retention indices are retention times normalised to adjacently eluting n-alkanes.
The range of the employed n-alkanes has to cover the expected retention time range of all possible target compounds. For monoterpenes and sesquiterpenes a range of C8 to C20 is usually suitable, oxygenated sesquiterpenes and diterpenes can require up to C26.
Definition of the retention index
RIx = 100 n0 + 100 (RTx – RTn0) / (RTn1 – RTn0)
with x the name of the target compound n0 n-alkane Cn0H2n0+2 directly eluting before x n1 n-alkane Cn1H2n1+2 directly eluting after x RT retention time (in any unit such as minutes, seconds, or scans) RI retention index (pure number without unit)
Examples of retention indices RI(n-decane) = 1000 RI(n-undecane) = 1100 RI(x) = 1050, for any x that elutes exactly in the middle between n-decane and n-undecane
Application of Retention Indices
Retention indices are established worldwide and used by a large number of scientist and laboratories. You may think of retention indices as a kind of natural property of a compound, which is in a complicated way related to the underlying gas/liquid equilibrium. Naturally, the RI is dependent on the kind of stationary phase and different stationary phases give rise to different RI of the same compound. Thus, the type of stationary phase should always be given when reporting retention indices. Beside this limitation, RI are system-independent, reliable and reproducible.
Retention indices are independent from
- delay of acquisition (absolute shift of time axis) has no influence on RI
- unit of time measurement (min, seconds, scans) has no influence on RI
- carrier gas pressure and flow rates have no influence if held constant during one measurement
- column length, column diameter and stationary film thickness have no influence on RI
- pre-columns have no or neglectible influence on the RI
- different isothermal temperatures have neglectible on RI
- different linear temperature ramps have neglectible on RI
Typical temperature programs composed of isothermal sections and linear temperature ramps give only slight to medium RI variations, particularly near the inflection points of the temperature profile. Best reproducibility is obtained if the temperature program consists of only one continuous temperature ramp. An inital isothermal period does no harm if the start of the temperature ramp is earlier then the retention time of the first actually required n-alkane standard. Likewise a final heating-off section does no harm if it starts later than the retention time of the last required n-alkane standard.
Again, different stationary phases can give entirely different retention indices. Using the same stationary phase is of utter importance for the successful employment of retention indices.
Reproducibility of retention indices
Generally, retention indices are reproducible on the same system with deviations equal or less than ± 2 RI, given that alkane standard measurements take place as required (see next chapter).
Generally, retention indices are reproducible between typical systems with deviations equal or less than ± 5 RI.
In the case of MassFinder's Terpenoids Library typical system means using a normal carrier gas velocity with respect to column length, a moderately fast continuous temperature gradient, and achieving overall good chromatographic resolution. Any commercially available polysiloxane column compatible with DB-1 (100% polydimethylsiloxane) should afford retention indices of approximately ± 5 RI or better when compared with our library's reference values. The more polar column DB-5 (95% polydimethylsiloxane-5%polydiphenylsiloxane) usually affords retention indices deviating less than ±10 RI for unpolar compounds and less than ±20 RI for polar compounds.
Limitations of retention index reproducibility
Retention indices might differ from reference values in cases such as
- overloading effects (visible as peak fronting or detector saturation)
- column activity (contaminated column, reactive parts, polar interactions)
- interconverting, reactive, or decomposing compounds (highly temperature dependent!)
Generally, all effects that might influence retention times under otherwise constant conditions will potentially influence retention indices as well. All processes that disturb the gas chromatographic separation or reduces or modifies column selectivity may cause significant deviations of retention indices.
Practical aspects of using retention indices
Internal vs. External Standards
Typical procedures for employment of retention indices does not use n-alkanes as internal standards. Routinely, all n alkanes necessary to cover your desired time range are mixed together in equal amounts in a single sample. This mixture is measured under your standard chromatographic conditions to obtain the retention times of all relevant n alkanes and the result table is called alkane pattern. There are commercial standard solutions available with n-alkanes of various ranges. The alkane pattern can usually be measured with a single injection.
As stated at the beginning of this document, retention times are valuable if identical chromatographic conditions are strictly maintained. A modern gas chromatograph reproduces retention times over several days or even weeks with satisfying accuracy and the alkane pattern has only to be measured again after system maintenance such as column shortening or pressure readjustment. Quality control regulations may require daily measurement of the alkane pattern, a single injection that increases reliability and validity of your GC data. Of course, the alkane pattern is different for each GC and the pattern has to be acquired for each system.
In summary, you may combine the advantages of short-term retention time reproducibility by separating the alkane pattern from the actual samples, and also profit from the use of retention indices due to inter-system and long-term compatibility. Even many years later and in a different laboratory you may reproduce your old retention indices while your retention times have no meaning anymore.
Should for whatever reason your application require to use the reference n-alkanes as internal standards, we recommend to restrict yourself to those n-alkanes absolutely necessary to calculate the retention indices of your analytes, i.e. those n-alkanes directly before, in between, and after the retention times of all primary target compounds. Thus, your run time will not be prolonged unnecessarily and the chromatogram will not be cluttered with useless peaks.
Using n-alkanes as internal standards is rarely necessary. Try to work with alkane patterns as external standard.
Are Other Reference Patterns Possible?
Naturally, the concept of retention indices requires a grid of several standard compounds distributed more or less equally over the whole possible retention time range. While n-alkanes are internationally established and you will easily find reference values based on n-alkanes, it is possible to use other compounds as standards. You just have to accept that the comparability is significantly reduced and you should have good reasons to deviate from the standard approach, e.g. a completely different stationary phase or very high elution temperatures with difficult to obtain n-alkanes. In such cases you can define your own set of standard compounds. We recommend choosing substances that are reasonably inert and highly temperature stable, whose retention times are equally distributed over your relevant time range and which time distance to each other is sufficiently short. Further, it should be very likely that the standard compounds will be available for many years to come and you need to document their identity exactly.
Retention Indices and Mass Spectrometry (GC/MS)
The challenge of identifying and assigning GC/MS peaks
A set of chemically different substances may give highly similar mass spectra, thus rendering it almost impossible to unambiguously distinguish the compounds from each other. The mass spectra shown below demonstrate this issue with three sesquiterpene hydrocarbons exhibiting pretty similar fragmentation patterns.
Introducing the second dimension...
In GC/MS technique the "GC" and the "MS" part are two independent experiments.
Usually substances with identical chromatographic retention times exhibit different mass spectra, or in other words, co-eluting peaks could only by pure chance give rise to identical mass spectra. Thus, mathematically spoken, the experimental values for retention times and mass spectra are different dimensions and independent properties. The chromatographic separation thus offers a very simple and powerful experimental value, the retention time or retention index, which can be used to distinguish compounds with identical mass spectra. Naturally, many compounds may give the same retention time, but usually these compounds exhibit different mass spectra. Only extremely rarely, two substances have by pure chance both identical retention times and identical mass spectra.
In summary, using retention indices in GC/MS does significantly increase the reliability of peak identifications by providing a second, independent experimental value. If you ignore retention times or retention indices in GC/MS and solely rely on mass spectral similarity, you waste precious information of an experiment already successfully performed.
Analysing your own data
Calculating retention indices from retention times is a tedious job and needs to be automated in order to reliably and rapidly employ retention indices. The GC/MS software MassFinder does support retention indices inherently and makes the usage of retention indices as well as comparing mass spectra and retention indices with libraries entries highly convenient.
Further reading on to set up MassFinder: Analysing your own data