Natural Synthesis of Hydrocarbons in
Alkaline Intrusive Rocks

Stefano Salvi Laboratoire de Géochimie, 38 rue des Trente-Six Ponts, F-31400 Toulouse, France

Anthony E. Williams-Jones Earth & Planetary Sciences, McGill University, 3450 University Street,

Montreal, Quebec, Canada, H3A-2A7

Carbonic gas is a very common constituent of the volatile phase in magmas and is dominated by oxidized species (i.e., CO and CO2) in both volcanic and intrusive environments (Gerlach, 1980; Roedder and Coombs, 1967; Holloway, 1976); these species are also thought to constitute the principal form of volatile carbon in the mantle (Pasteris and Wanamaker, 1988; Bergman and Dubessy, 1984). Reduced carbonic species have been reported only from inclusion fluids in silica-undersaturated alkaline rocks (Petersilie, 1963; Sobolev et al., 1974; Konnerup-Madsen et al., 1979) and more recently from inclusion fluids in peralkaline granites (Salvi and Williams-Jones, 1992). These fluids are rich in CH4, contain heavier alkanes (C2 to C5 and higher) and H2, and may contain significant N2, CO2, CO, and aromatic hydrocarbons.

The origin of reduced fluids in igneous rocks is still a matter of considerable debate. Some authors postulate that these fluids represent organic carbon, graphite or carbonate assimilated by the magma (Konnerup-Madsen et al., 1988), and others that they originate from inorganic gases produced by the magma (Petersilie and Sørensen, 1970; Kogarko et al., 1987). Supporters of the latter hypothesis envisage an initial separation of a CO2-bearing fluid at relatively reduced conditions, which subsequently transforms to hydrocarbons on cooling. One study (Petersilie and Sørensen, 1970), in order to explain the existence of heavier hydrocarbons, suggested that they might form by reactions between CO and H2 or H2O catalyzed by nepheline, but provided few details.

In the Strange Lake peralkaline granite, Quebec/ Labrador, pegmatites and adjacent altered granites contain immiscible aqueous and carbonic fluid inclusions, interpreted to have originated from the magma (Salvi and Williams-Jones, 1992). Gas chromatographic analysis of the gaseous fraction in these inclusions revealed the presence of methane, hydrogen, and heavier hydrocarbons (in decreasing order of abundance, C2H6, C3H8, n-C4H10, n-C5H12, C2H2,
i-C4H10, C2H4, i-C5H12, n-C6H14, i-C6H14 and neo-C6H14). Thermodynamic calculations in the system C-O-H show that the hydrocarbons could not have co-existed at equilibrium in the proportions measured, either prior to or after entrapment, nor that they could have formed by reaction of the aqueous fluid with graphite. Instead, we propose that the higher hydrocarbon species in the Strange Lake fluid inclusions were produced by the disequilibrium reactions:

in a very similar manner to the Fischer-Tropsch synthesis, which is used industrially to convert coal to petroleum in the presence of a Group VII metal or metal oxide catalyst (Anderson, 1984). This interpretation is supported by the fact that the ratios of the mole fractions of pairs of progressively longer chain hydrocarbons obey the Schulz-Flory law, which states that in a Fischer-Tropsch synthesis the molecular ratios of hydrocarbons with successively higher carbon numbers are constant (Cn+1/Cn = Cn+2/Cn+1).

Although the magma is the obvious source for the CO2 or CO needed to produce the hydrocarbons, equilibrium calculations indicate that the exsolved gas will contain negligible H2 at the temperature and fO2 conditions interpreted for degassing (500ƒ to 600ƒC; 2 to 3 log units below QFM). We propose that H2 was produced during subsolidus alteration of arfvedsonite to aegirine as a result of interaction of the granite with fluids released from the pegmatite-forming magma:

3Na3FeII4FeIIISi8O22(OH)2 + 2H2O =


9NaFeIIISi2O6 + 2Fe3O4 + 6SiO2 + 5H2

aegirine magnetite quartz

and that magnetite produced by this alteration catalysed the hydrocarbon synthesis.


Anderson, R.B., The Fischer-Tropsch Synthesis Academic Press, New York, (1984).

Bergman, S.C. & Dubessy, J., Contrib. Mineral. Petrol. 85, 1-13 (1984).

Gerlach, T.M., J. Volcanol. Geoth. Res. 7, 295-317 (1980).

Holloway, J.R. Geol. Soc. Am. Bull. 87, 1513-1518 (1976).

Kogarko, L.N., Kosztolanyi, Ch. & Ryabchikov, I.D., Geochem. Int. 24, 20-27 (1987).

Konnerup-Madsen, J., Kreulen, R. & Rose-Hansen, J., Bull. Minéral. 111, 567-576 (1988).

Konnerup-Madsen, J., Larsen, E. & Rose-Hansen, J., Bull. Minéral. 102, 642-653 (1979).

Pasteris J.D. & Wanamaker, B.J., Am. Mineral., 73, 1074-1088 (1988).

Petersilie, I.A., In Chemistry of the Earth's Crust (Vinogradov, A.P., ed.) 1, 178-184 (1963)

Petersilie, I.A. & Sørensen, H., Lithos 3, 59-76 (1970).

Roedder, E. & Coombs, D.S., J. Petrol. 8, 417-451 (1967).

Salvi, S. & Williams-Jones, A.E., Eur. J. Mineral. 4, 1155-1174 (1992).

Sobolev, V.S., Bazarova T.Y. & Kostyuk, V.P., In The Alkaline Rocks Sørensen, H. (ed.) . Wiley, New York, 389-401 (1974).