Preparation of Nitrogen-Containing Derivatives for Mass Spectrometry of Fatty Acids

The methodology, including practical details, is given here for preparation of the nitrogen-containing derivatives of fatty acids, i.e. 3-pyridylcarbinol ('picolinyl') esters, 4,4-dimethyloxazoline (DMOX) derivatives and pyrrolidides, which are most useful for mass spectrometric analysis. Preparation of methyl ester derivatives is described on a separate webpage. Other useful derivatization techniques, e.g. hydrogenation, deuteration, etc., are described in the pages dealing with Mass spectra of methyl esters - further derivatization. I must also recommend my book [1].



For some derivatives, it can be better to prepare the free acids first by hydrolysis of the lipids with dilute aqueous ethanolic alkali as described here [1].

A solution of 0.1M potassium hydroxide in 90% aqueous ethanol is prepared immediately before use by adding 1 volume of standard 1M aqueous potassium hydroxide to 9 volumes of ethanol. To the lipid sample (up to 6 mg) in a stoppered centrifuge tube is added 0.1M potassium hydroxide in 90% aqueous ethanol (0.25 mL per mg sample). Leave at 50°C in a water bath or heating block for 3 hours. Remove from water bath and add 2M hydrochloric acid (0.05 mL per mg sample) using a Pasteur pipette. Add isohexane (3 mL), diethyl ether (3 mL) and distilled water (2 mL). Shake thoroughly and, if necessary (if there are not two distinct layers and/or an emulsion is formed), centrifuge for 5 min (about 400 rpm). Put upper organic layer through a short (3 cm) column of anhydrous sodium sulfate (prepared in a Pasteur pipette, plugged with a small piece of cotton wool, and pre-washed with 3 mL isohexane-diethyl ether (1:1, by vol.) prior to use) to dry. Collect in a test tube. Repeat extraction of lower aqueous layer with isohexane-diethyl ether (3 mL, 1:1, by vol.) and put the organic layer through the sodium sulfate column. Wash sample through the column into the tube with more isohexane-diethyl ether (3 mL). Take sample to dryness in a gentle stream of nitrogen on a heating block at 30°C.


3-Pyridylcarbinol ('picolinyl') Esters

3-Pyridylcarbinol esters, commonly but incorrectly known as ‘picolinyl’ esters, are most easily prepared in good yield from the free fatty acids prepared as above following activation by conversion to the acid chlorides. Reaction with thionyl chloride is rapid and efficient [2], but the milder method of reaction with oxalyl chloride overnight is also suitable [3]. Dry solvents and fresh reagents are required because the reactions are sensitive to moisture.

Preparation of 3-pyridylcarbinol (picolinyl) esters via acid chlorides

The reactions are carried out in 0.3 ml screw cap vials. The acid (20 µg) is heated with thionyl chloride (20 µL) for 10 min. at 100ºC. Excess reagent is evaporated in a stream of nitrogen, and a solution of 20% 3-(hydroxy-methyl)pyridine in acetonitrile (10 µL) is added; the mixture heated for 1 min. at 100ºC. An aliquot can be injected directly onto the GC column.

A mild quantitative method developed for derivatizing sensitive polyunsaturated fatty acids containing epoxyl groups involving an imidazolide intermediate is a useful alternative [4]. It is simple and rapid, involving brief reactions with first carbonyldiimidazole then 3-(hydroxymethyl)pyridine and a catalyst.

Preparation of 3-pyridylcarbinol (picolinyl) esters via imidazole derivatives

The reaction, as detailed below, is moisture sensitive and all reagents should be protected from moisture (i.e. by storing in a the presence of a desiccant). Fresh carbonyldiimidazole solution should be prepared each time a set of samples is derivatized.

The derivatizing reagent is prepared by dissolving 3-(hydroxymethyl)pyridine (900 µL) in dichloromethane (9 mL) and triethylamine (9 mL). It can be used for up to one month if kept in a sealed tube (with PTFE-lined screw top) with a little added anhydrous sodium sulfate as drying agent in a refrigerator. The free fatty acid sample (up to 1 mg), in a centrifuge tube, is dissolved in dichloromethane (100 µL) and a solution of 1,1'-carbonyldiimidazole (100 µL, 10 mg/mL in dichloromethane, freshly prepared) is added. Leave 1 min. (not longer) at room temperature before proceeding to the next step. Add the 3-pyridylcarbinol reagent (200 µL). Agitate and leave for 10 min. at 37°C. At the end of this time add acetic acid (25 µL). Take to dryness on a heating block at 30°C with a gentle stream of nitrogen. Add isohexane (5 mL) and distilled water (2 mL). Shake thoroughly,then put upper isohexane layer through a short (4 cm) column of anhydrous sodium sulfate in a Pasteur pipette (with a small cotton wool plug and pre-washed with isohexane) to dry. Re-extract the aqueous layer with fresh isohexane (2 mL), and pass through the column also. If necessary, a little crystalline potassium chloride can be added to break emulsions. Finally, wash through column with isohexane (1 mL) and evaporate the combined isohexane eluents. Dissolve sample in an appropriate amount of isohexane for GC and GC-MS analyses.

It is also possible to prepare 3-pyridylcarbinol esters directly from intact lipids or methyl esters by transesterification with 3-(hydroxymethyl)pyridine with potassium tert-butoxide in tetrahydrofuran as catalyst [5]. This reaction may indeed be useful for the preparation of a variety of different ester derivatives by substituting the appropriate alcohol. The reagent must be very dry, otherwise an irreversible hydrolysis reaction can occur. It is best to make a fresh solution of potassium tert-butoxide in dry tetrahydrofuran each time it is required.

Preparation of 3-pyridylcarbinol (picolinyl) esters by base-catalysed transesterification

A solution of potassium tert-butoxide in tetrahydrofuran (0.1 mL, 1.0 M) is added to 3-pyridylcarbinol (0.2 mL). After mixing, the lipid sample (up to 10 mg) in dry dichloromethane (1 mL) is added, and the solution is held at 40°C for 30 min in a closed vial. After cooling to room temperature, water (2 mL) and hexane (4 mL) are added, and the organic phase is collected, dried over anhydrous sodium sulfate, and evaporated. The sample is dissolved in isohexane for GC-MS analysis.

3-Pyridylcarbinol esters can be converted back to methyl esters by conventional acidic or basic transesterification procedures.


Dimethyloxazoline (DMOX) Derivatives

DMOX derivatives are prepared in a simple one-pot reaction by reaction of lipids with 2-amino-2-methyl-1-propanol in a nitrogen atmosphere at 180°C (a minimum of 2 hours for free acids, or up to 18 hours for methyl esters and intact lipids [6].

Preparation of DMOX derivatives by the one-pot method

To the lipid sample (up to 2 mg) in a test tube is added 2-amino-2-methyl-1-propanol (0.25 g). The vessel is flushed with nitrogen, stoppered, and placed in a heating block, at 190ºC overnight. Flush the tube with nitrogen when the tube is up to temperature to help to eliminate moisture and minimize autoxidation. A little glycerol added to the well of the heating block ensures good contact and that the correct temperature is maintained. Next day, remove the test tube from the heating block, allow to cool to room temperature and wash off any glycerol on the external surface with water. Add diethyl ether-isohexane (1:1, v/v; 5 mL) to the tube, washing down the internal surface, followed by water (5 mL). Shake thoroughly and allow layers to settle. If necessary, addition of a little sodium or potassium chloride will usually break up any interfacial layers. Transfer the organic layer with a pasteur pipette to a fresh test-tube, and re-extract the aqueous layer with fresh solvent (2 mL). Add distilled water (3 mL) to the combined solvent layers, shake and allow to settle. Transfer the solvent layer to a fresh tube with a pasteur pipette and add anhydrous sodium sulfate (about 1 cm). Leave for one hour with occasional vigorous shaking, then pass the solvent layer through a short (3 cm) column of anhydrous sodium sulfate (prepared in a pasteur pipette, plugged with a small piece of cotton wool, and pre-washed with ca. 3 mL isohexane prior to use) to dry, into a fresh test tube. Wash sample from the sodium sulfate tube through column with isohexane (2 mL). Take sample to dryness in a gentle stream of nitrogen on a heating block at 30ºC. Dissolve the sample in an appropriate amount of isohexane for GC and GC-MS analyses. Adding a few crystals of anhydrous sodium sulfate stabilizes the derivative for storage.

Unfortunately, we have often observed incomplete reaction and appearance of the uncyclized intermediate, which elutes later from GC columns (and gives a mass spectrum almost identical to that of the required derivative). It is important to keep the product and reaction mixtures as dry as possible, otherwise traces of moisture can cause ring opening, even on storage in inert solvents at low temperatures [1]. For the same reason, it is not possible to purify DMOX derivatives by adsorption chromatography.

The prolonged high temperature required for the preparation of DMOX derivatives gives cause for concern, and there must be some risk to polyunsaturated fatty acids or any other compound with a labile functional group. For example, trans-3-hexadecenoic acid, common in plant photosynthetic tissue, was found to have isomerized largely to cis-2-hexadecenoic acid during the reaction [7], and crepenynic acid was found to undergo a cyclization reaction [8]. We have also observed that linoleic acid can be partially isomerized to conjugated isomers ('CLA') if the reaction is carried out carelessly.

An alternative two-step reaction has been described that may be safer in some circumstances [8]. It involves preparing the simple amide via the acid chloride, followed by cyclization with trifluoroacetic anhydride. As with 3-pyridylcarbinol esters, I prefer to prepare DMOX derivatives via the free acids rather than from intact lipids but this is not always practical or convenient.

Preparation of DMOX derivatives by a two-stage method

The fatty acid must first be converted to the acid chloride as in the methods for preparation of 3-pyridylcarbinol esters via the acid chloride. This is used immediately for the preparation of the required derivatives. A solution of 2-amino-2-methylpropanol in dichloromethane (10 mg/mL) is prepared and stored over anhydrous sodium sulfate. It is stable in the refrigerator for about 1 month. This solution (0.5 mL) is added to the freshly prepared acid chloride, cooled in an ice bath, and the mixture is then left to warm up to room temperature for an hour. The solvent is removed in a stream of nitrogen, and trifluoroacetic anhydride (0.5 mL) is added. After 1 hour on a heating block at 50ºC, or 3 hours at room temperature, the reagent is removed in a stream of nitrogen. Isohexane (5 mL) is added followed by water (2 mL), and the product is obtained as in the previous reaction, taking care to dry the product thoroughly.

A further two-step procedure that is suitable for the preparation of both DMOX and other simple oxazoline derivatives has been described, and the GC-MS properties of the latter appear excellent [9]. However, few published spectra are available. Of the many alternative nitrogen-containing derivatives for mass spectrometry that have been described in the literature, the latter appear to me to be the only ones that might be worth pursuing. Similarly, a new mild procedure for preparing DMOX derivatives and pyrrolidides from free acids using Deoxo-FluorTM reagent to activate the fatty acids, which should also be applicable to pyridylcarbinol esters, has been described [10,11]. We have yet to try it out so would be interested in learning of the experience of others.

DMOX derivatives can be converted back to methyl esters by methylation procedures used for sphingolipids, e.g. by reaction with methanol containing hydrochloric acid [1].



Pyrrolidides are prepared simply by reacting methyl esters of fatty acids with pyrrolidine, in the presence of acetic acid [12].


The fatty acid methyl ester (up to 10 mg) is dissolved in freshly distilled pyrrolidine (1 mL), acetic acid (0.1 mL) is added, and the mixture is heated at 100ºC for 1 hour. Excess pyrrolidine is blown off in a stream of nitrogen at 50ºC, and then the residue is taken up in hexane-diethyl ether (1:1, v/v; 8 mL) and is washed three times with water (4 mL portions). After drying over anhydrous sodium sulfate, the required product is obtained on evaporation of the solvent.

The reaction takes place under basic conditions and is relatively mild so may be safer than for DMOX derivatives with fatty acids containing labile functional groups. Alternative procedures are also available to prepare pyrrolidides directly from triacylglycerols and presumably other glycerolipids [13] or from free fatty acids [10,14,15], the latter in a mild one-pot reaction.



  1. Christie, W.W. and Han, X. Lipid Analysis - Isolation, Separation, Identification and Lipidomic Analysis (4th edition), 446 pages (Oily Press, Woodhead Publishing and now Elsevier) (2010) - see Science Direct.
  2. Harvey, D.J. Picolinyl esters as derivatives for the structural determination of long chain branched and unsaturated fatty acids. Biomed. Mass Spectrom., 9, 33-38 (1982) (DOI: 10.1002/bms.1200090107).
  3. Mattson, F.H. and Volpenhein, R.A. Synthesis and properties of glycerides. J. Lipid Res., 3, 281-296 (1962).
  4. Balazy, M. and Nies, A.S. Characterization of epoxides of polyunsaturated fatty acids by mass spectrometry via 3-pyridinylmethyl esters. Biomed. Environ. Mass Spectrom., 18, 328-336 (1989) (DOI: 10.1002/bms.1200180508).
  5. Destaillats, F. and Angers, P. One-step methodology for the synthesis of FA picolinyl esters from intact lipids. J. Am. Oil Chem. Soc., 79, 253-256 (2002) (DOI: 10.1007/s11746-002-0469-7).
  6. Fay, L. and Richli, U. Location of double bonds in polyunsaturated fatty acids by gas chromatography-mass spectrometry after 4,4-dimethyloxazoline derivatization. J. Chromatogr. A, 541, 89-98 (1991) (DOI: 10.1016/S0021-9673(01)95986-2).
  7. Lamberto, M. and Ackman, R.G. Positional isomerization of trans-3-hexadecenoic acid employing 2-amino-2-methylpropanol as a derivatizing agent for double bond location by GC/MS. Anal. Biochem., 230, 224-228 (1995) (DOI: 10.1006/abio.1995.1467).
  8. Christie, W.W. Mass spectrometry of fatty acids with methylene-interrupted ene-yne systems. Chem. Phys. Lipids, 94, 35-41 (1998) (DOI: 10.1016/S0009-3084(98)00016-4).
  9. Kuklev, D.V. and Smith, W.L. A procedure for preparing oxazolines of highly unsaturated fatty acids to determine double bond positions by mass spectrometry. J. Lipid Res., 44, 1060-1066 (2003) (DOI: 10.1194/jlr.D200046-JLR200).
  10. Kangani, C.O. and Kelley, D.E. One pot direct synthesis of amides or oxazolines from carboxylic acids using Deoxo-Fluor reagent. Tetrahedron Letts., 46, 8917-8920 (2005) (DOI: 10.1016/j.tetlet.2005.10.068).
  11. Kangani, C.O., Kelley, D.E. and Evans, R.W. Synthesis and mass spectrometry of benzoxazoline, dimethyloxazoline and 4-phenyloxazoline derivatives of polyunsaturated fatty acids. Rapid Commun. Mass Spectrom., 21, 2129-2136 (2007) (DOI: 10.1002/rcm.3071).
  12. Andersson, B.A. and Holman, R.T. Pyrrolidides for mass spectrometric determination of the position of double bonds in monounsaturated fatty acids. Lipids, 9, 185-190 (1974) (DOI: 10.1007/BF02532690).
  13. Vetter, W. and Walther, W. Pyrrolidides as derivatives for the determination of the fatty acids of triacylglycerols by gas-chromatography. J. Chromatogr. A, 686, 149-154 (1994) (DOI: 10.1016/0021-9673(94)00079-4).
  14. Vetter, W. and Walther, W. Preparation of pyrrolidides from fatty acids via trimethylsilyl esters for GC-MS analysis. J. Chromatogr. A, 513, 405-407 (1990) (DOI: 10.1016/S0021-9673(01)89466-8).
  15. Kangani, C.O., Kelley, D.E. and DeLany, J.P. New method for GC/FID and GC-C-IRMS analysis of plasma free fatty acid concentration and isotopic enrichment. J. Chromatogr. B, 873, 95-101 (2008) (DOI: 10.1016/j.jchromb.2008.08.009).

Updated May 5, 2014