A Beginner's Guide to Mass Spectrometry of Fatty Acids: Part 1
At its simplest, mass spectrometry (MS) is a technique in which organic molecules are bombarded by electrons or other ionic species causing them to ionize and fragment. We use the fragmentation patterns as a means of identification. If you pick up a journal dealing with mass spectrometry, there will be a host of technical terms dealing with methods of ionization especially, such as "electrospray ionization", "chemical ionization", and "fast-atom bombardment", but in the more basic mass spectrometers linked to gas chromatography (GC-MS), electron impact ionization is encountered most often. It is the one most relevant to this specific topic and is the only one considered here. The various ionic species produced from a given organic compound by electron impact are separated according to mass (strictly speaking mass/charge (m/z) ratio in which z = 1) in a magnetic field, and a spectrum is obtained that in effect is a bar diagram showing the masses of the fragment ions and their abundances relative to the most abundant ion (base ion), which is usually normalized to 100% for comparison and publication purposes.
In this and the next article, I discuss the basic principles only of identification of fatty acid structures by mass spectrometry. More detailed information is available in the other articles in this part of the website, and two review articles can be recommended [1,2].
Jigsaw Puzzles and Bricks
Interpretation of a mass spectrum is often compared to doing a jigsaw puzzle. We have many different pieces or fragments and we have to try to put them together in a sensible way to find the picture or to describe the molecule. I prefer an analogy, in which mass spectrometry is compared to demolishing and re-assembling a brick wall. If we use a sledgehammer to create a pile of rubble, we will know the total mass present but reassembling the wall is impossible. If we can take the wall apart cleanly a few bricks at a time, it will be possible to reassemble it easily. We will know the correct dimensions and where any door or window should be placed. With a fatty acid derivative, in comparison, we need to confirm that it is indeed a fatty acid, determine the molecular weight and then locate any double bonds or other functional groups.
Methyl esters are the derivative of choice in most analytical applications of lipids. However, when we subject them to electron impact in a mass spectrometer, the charged aliphatic chain breaks up into an indeterminate number of fragments, so a branch-point or ring structure may not be identified easily. For example, a double bond can carry the charge and becomes mobile; it moves up and down the chain, so its original position cannot be determined reliably. This is especially troublesome with mono- and dienoic fatty acids. Using the above analogy, we have taken a sledgehammer to break up the molecule. Yet all is not lost, as the molecular weight of the fatty acid ester is obtained from the molecular ion, and therefore the number of carbon, hydrogen and oxygen atoms. This tells us if the fatty acid is saturated or unsaturated, and often an experienced eye can tell if there is a branch point or other substituent.
For example, the mass spectrum of methyl oleate is illustrated (Figure 1). The molecular ion is at m/z = 296, i.e. two less than for methyl stearate. We usually lose one or two 'bricks' neatly. An ion representing loss of 32 mass units (methanol from the ester moiety) and another of 74 units (both m/z = 74 per se and [M-74]+ at m/z = 222) from a rearrangement involving the ester moiety confirm that we really do have a methyl ester. However there are no ions that help us to determine the position or configuration (cis/trans) of the double bond.
Strictly speaking, as the molecular weight is two less than for a fully saturated fatty acid derivative, we could have either a double bond or a cyclic structure present, but the correct assignment is usually possible from simple experiments, for example by attempting hydrogenation.
The other information we should have as an aid to identification is the relative retention time of the fatty acid derivative (or equivalent chain-length value) on the GC column of the GC-MS system. This together with the molecular weight data will suggest that some structures are more plausible than others.
Double Bond Location in Polyunsaturated Fatty Acid Methyl Esters
More often than not we wish to know where double bonds are located in an aliphatic chain, and the mass spectrum of the methyl esters are not always of assistance. Although we cannot do this directly with monoenes and dienes, it is fortunate that with the conventional series of polyunsaturated fatty acids, i.e. with three or more methylene-interrupted double bonds, we have some useful characteristic ions. These can be seen in the mass spectrum of methyl 6,9,12-octadecatrienoate (γ-linolenate) illustrated in Figure 2 as an example.
First, the molecular ion at m/z = 292 is 6 units less than for methyl stearate (18:0), which tells us that there are three double bonds. The ion at m/z = 74 (the so-called McLafferty ion) confirms that we do indeed have a methyl ester. These ions tend to be much smaller than in spectra from more saturated esters. Ions from the hydrocarbon part of the molecule of general formula [CnH2n-5]+ tend to dominate the spectrum with the ion at m/z = 79 as the base peak, but these tell us little about the detailed structure.
However, an ion at m/z = 150 (‘omega’ ion) formed by a fragmentation at the terminal end of the molecule is characteristic of all polyunsaturated fatty acids from the n-6 biosynthetic series . Also, there is a fragment (‘alpha’ ion) from the carboxyl end of the molecule at m/z =194, which completes the identification, although the position of this ion will vary according to the position of the first double bond in fatty acids of the n-6 series. The latter ion was defined in a paper by Brauner and coworkers , which until recently had been overlooked by most experts in this field.
In mass spectra of methyl esters of fatty acids of the n-3 series, the omega ion stands out at m/z = 108 (and that at m/z = 150 is negligible). For the minor (n-9) and (n-4) biosynthetic families the relevant ions are at m/z = 192 and 122, respectively. It should be stressed that these ions are not infallible guides, and are not reliable when the double bonds are not methylene interrupted.
That said, it is surprising how often mass spectra from methyl esters of polyunsaturated fatty acids are sufficiently distinctive to be considered as ‘fingerprints’, which can be identified by comparison with the spectra of standards. Unfortunately, when there are five or more double bonds, the molecular ion can be of low abundance and difficult to locate.
Chemical Derivatization for Double Bond Location
If we want to have definitive information on double bond positions, we have to get away from the sledgehammer approach. There are two ways this can be done. One is to prepare particular nitrogen-containing derivatives of fatty acids, and this will be the topic of the Part 2 of my beginner’s guide. Alternatively, with monoenoic fatty acids especially, it is possible to "fix" the double bonds by reacting them with appropriate reagents to give chemical derivatives that give distinctive fragmentations in the mass spectrometer.
A host of such derivatives have been described, usually involving oxygenation followed by further derivatization, but most of these fell by the wayside when dimethyl disulfide adducts were described . This is simple single-step (if odoriferous) reaction involving reaction of dimethyl disulfide with an unsaturated ester in the presence of iodine as catalyst. The nature of the reaction is shown in Figure 3.
For example with methyl oleate, the molecular weight increases substantially (from 296 to 390) but this is still in a comfortable range for GC analysis. The mass spectrum gives a good molecular ion but the most abundant ions represent cleavage at carbon atoms that were originally linked by the double bond (m/z = 173 and 217 as shown), and this is located unequivocally. This technique has now been applied to identify many monoenoic fatty acids (and other aliphatic compounds) in natural samples in a number of laboratories. (This is the best test of any method - often a new procedure is described with one or two model compounds and then falls into oblivion).
The technique is less straight-forward if there is more than one double bond in the molecule, as bis-adducts are only formed quantitatively if there are more than three carbons between double bonds, for example. However, the reaction can be used usefully even with methylene-interrupted dienes, such as linoleic acid, if the reaction is carried out under mild conditions since two mono-adducts (one double bond derivatized and one not) are then formed and these are easily separable by GC for analysis by MS . (There is a now more substantial article on this aspect of the topic on the website here..)
Such methods add an extra complication to analysis, and it would obviously be much easier if a single derivative could be used for all purposes, as discussed in Part 2.
Be Wary of Computerized Identification Techniques
Most GC-MS systems are sold with software that enables comparison of spectra of unknowns with a library of spectra of standards. Beginners should be very careful of how this facility is used for a number of reasons. While the library may contain 100,000 spectra or more, this is still a small number in comparison to the number of organic compounds in general and fatty acid derivatives in particular that might be encountered. Then, the identification is made from the intensities of a relatively small number of ions, and ion intensities can be rather variable, depending on instrumental factors. A comparison with the spectrum of methyl oleate, for example, may suggest a number of positional or configurational isomers, all with a confidence of 99%. The computer does not recognise that isomeric methyl octadecenoates of this kind have spectra that are identical for all practical purposes. Analysts should have a sufficient understanding of the principles of mass spectrometry to know when to ignore the computer and make informed decisions.
- Christie, W.W. Structural analysis of fatty acids. In Advances in Lipid Methodology - Four, pp. 119-169 (edited by W.W. Christie, Oily Press, Dundee) (1997).
- Christie, W.W. Gas chromatography-mass spectrometry methods for structural analysis of fatty acids. Lipids, 33, 343-353 (1998) (DOI: 10.1007/s11745-998-0214-x).
- Fellenberg, A.J., Johnson, D.W., Poulos, A. and Sharp, P. Simple mass spectrometric differentiation of the n-3, n-6 and n-9 series of methylene interrupted polyenoic acids.Biomed. Environ. Mass Spectrom., 14, 127-130 (1987) (DOI: 10.1002/bms.1200140306).
- Brauner, A., Budzikiewicz, H. and Boland, W. Studies in chemical ionization mass spectrometry. 5. Localization of homoconjugated triene and tetraene units in aliphatic compounds. Org. Mass Spectrom., 17, 161-164 (1982) (DOI: 10.1002/oms.1210170403).
- Francis, G.W. Alkylthiolation for the determination of double bond position in unsaturated fatty acid esters. Chem. Phys. Lipids, 29, 369-374 (1981) (DOI: 10.1016/0009-3084(81)90070-0).
- Yamamoto, K., Shibahara, A., Nakayama, T. and Kajimoto, G. Determination of double bond positions in methylene-interrupted dienoic fatty acids by GC-MS as their dimethyl disulfide adducts. Chem. Phys. Lipids, 60, 39-50 (1991) (DOI: 10.1016/0009-3084(91)90013-2).
The second part of this beginner's guide to mass spectrometry of fatty acids can be accessed here..
Updated: February 19th, 2014