Retardation by Water Pressure of Hydrocarbon Generation Reactions in Geological Basins: An Experimental Investigation

A. D. Carr^1,5, C. Uguna², C. E. Snape², W. Meredith², I. C. Scotchman³, and R. C. Davis⁴
¹Advanced Geochemical Systems Ltd., 1 Towles Fields, Burton-on-the-Wolds, U.K
²Nottingham Fuel & Energy Centre, School of Chemical, Environmental and Mining Engineering, University of Nottingham, U.K
³Statoil (UK) Ltd., Statoil House, 11A Regent Street, London, U.K
⁴Woodside Energy (USA) inc., 5151 St. Felipe, Houston, Texas, USA
⁵British Geological Survey, Keyworth, Nottingham, UK

Hydrocarbon generation is generally modelled as primarily a temperature dependent process, with time being of much lower importance. Pressure as a chemical parameter controlling the reaction rates in petroleum basins has largely been ignored, even though physical chemical theory is explicit in stating that endothermic volume expansion reactions, e.g. maturation, hydrocarbon generation and oil cracking, are controlled by the temperature and pressure of the system (Atkins and de Paula, 2002). The absence of pressure controlling reaction rates is based on experimental investigations which showed that pressure has either no or only minor control on the maturation, hydrocarbon generation and oil cracking reaction rates, when water (excluding water derived by dehydroxylation) is absent from the reaction system (see Michels et al., 1995). This work has resulted in the widespread use of Arrhenius (non-equilibrium) kinetic models for maturation and hydrocarbon generation. Other studies however have shown a significant pressure effect when water is used as the pressure medium, and that water pressure may be an important factor in high pressure, low temperature (relative to the temperatures used in laboratory pyrolysis) environments such as geological basins (Michels et al., 1995; Carr, in press). This study will present the results from an extended study on the pyrolysis of a type II oil-prone Kimmeridge Clay source rock to show that temperature and water pressure together control the extent of the reactions in geological basins.

Why should water pressure determine the extent of hydrocarbon generation in either laboratory experiments or in geological basins? The conventional geochemical model considers a kerogen molecule as being activated by increasing temperature, such that the molecule expands and reaction occurs as the thermal barrier provided by the activation energy (Ea) is exceeded. However the expansion does not occur in vapour-space as used by most geochemical pyrolysis methods, e.g. Rock-Eval and MSSV, but the reacting molecule expands against the force exerted by surrounding environment, which in the case of a subsurface geological basin is mainly water pressure in the rock porosity and kerogen microporosity. As the reacting molecule expands it transfers thermal energy from the kerogen into pV work against the water pressure as the water is pushed away. Transition state theory shows that this pV work produces an increase in the activation energy as given by Eqns. 1 and 2. At a given temperature the higher the water pressure the greater will be the retardation of hydrocarbon generation, since thermal energy that could be used to overcome the activation energy barrier in gas-phase pyrolysis experiment, now has to be used to force the pressurised water away from the activated complex.

Ea = H^‡ + R.T (1)

/\H^‡ = /\U^‡ + /\pV^‡ (2)

where /\H^‡ is the enthalpy of activation, /\U^‡ is the internal energy at activation and /\pV^‡ is the work against the force of the water that occurs as the volume of the activated complex increases.

Using a pressure vessel rated up to 500 bar (ca. 7000 psi) at 350^oC, and up to 450 bar (ca. 6500 psi) at 420^oC, a Type II Kimmeridge Clay from Dorset, UK has been pyrolysed under anhydrous, hydrous and water pressure. The source rocks were heated for between 6 and 24 hours under atmospheric pressures with inert gas, steam and (distilled) liquid water pressure to examine the effects of the phase and pressure on vitrinite reflectance and hydrocarbon (bitumen/oil and gas) generation.

Carr et al. (in press) and Uguna et al. (in preparation) have both shown that 500 bar of water pressure at 500^oC retards hydrocarbon and more specifically gas generation from the Kimmeridge Clay and two coals (Longannet, UK and Svalbard, Norway) (Fig. 1). Although these findings are significant, the amount of reaction or the degree of transformation is relatively low whereas high transformations are required for any kinetic model derivation, and the temperatures were therefore increased to 420^oC. At temperatures of 374^oC water becomes supercritical, and we were concerned that supercritical water may not produce the same effect as liquid water. Non-corrosive salts were used in an attempt to prevent the water from becoming supercritical at temperatures above 400^oC, but our PVT work showed that this was not the case. Regardless of the physical state of the water, the structural limitations of the pressure vessel mean that at temperatures of 420^oC the maximum pressure that can be used is 450 bar, and the experiments were run under these maximum T-p conditions.

The results at 350^oC for 24 hours show that apart from bitumen generation from the Kimmeridge Clay, the bitumen generation from the two coals and the gas generation in all cases reaches a maximum under hydrous conditions (both liquid water and vapour present in the vessel). At 500 bar the bitumen yields obtained from the two coals are slightly reduced compared with the 20 ml hydrous values. However the gas yields are all reduced at 500 bar relative to the 10 and 20 ml hydrous values.

The results from the 380^o and 420^oC experiments at 6, 12 and 24 hrs for Kimmeridge Clay are shown in Fig. 2. The bitumen yield decreases and gas yield increases as the temperature increases. The increase in gas yield with temperature occurs due to the combined effect of the conversion of some of the bitumen and kerogen into gas. At each temperature both the bitumen and gas yields also increase with increasing time from 6 to 12 hours, although this effect is less apparent when comparing the 12 and 24 hours data. At temperatures of 310 and 320^oC, the addition of water in a hydrous experiment (part vapour, part liquid water, system pressure <160 bar) produced an initial increase in both bitumen and gas yields, and was attributed to the beneficial effect of water in the hydrous experiments (Carr et al., in press). At 380^oC the bitumen/oil yield appears to reach a maximum at a pressure between 350 and 400 bar, while at 420^oC the bitumen yield increases from 60 bar to 450 bar, suggesting that the retardation effect of pressure is being moved to a higher pressure than 500 bar by the use of high temperature (420^oC) water. The increase in bitumen/oil yield at 380^oC and 420^oC going from 60 bar to 450 bar is due to pressure retarding the cracking of the generated bitumen to lighter oil, which means pressure is preserving the bitumen/oil. At 380^oC, pressure slightly retarded gas generation, while at 420^oC gas generation was not retarded by pressure suggesting that the retardation effect of pressure on gas generation is being moved to higher pressures than 500 bar. Vitrinite reflectance measurements show that the maturity increases with both increasing temperature, pressure and time, although the result show that at 400 to 500 bar, the reflectance values are always slightly lower than the values obtained under hydrous conditions.

The implications for hydrocarbon generation in geological basins are potentially extremely significant. The source rocks in many geological basins are generally at lower temperatures than 200^oC, but higher pressures than 500 bar. Laboratory experimentation in order to produce sufficient transformation required for kinetic modelling in the short time intervals must use temperatures in excess of 350^oC, and the yields in such experiments are going to reflect the temperatures used. However, the lower temperatures and higher pressures of source rocks mean that hydrocarbon generation is controlled predominantly by the thermal and pressure histories of the basins.

References

Carr, A.D., Snape, C.E., Meredith, W., Uguna, C., Scotchman, I.C.and Davis, R.C. in press. The effect of water pressure on hydrocarbon generation reactions: some inferences from laboratory experiments. Accepted for Petroleum Geoscience.

Dalla Torre, M., Mahlmann, R.F. and Ernst, W.G. (1997). Experimental study on the pressure dependence of vitrinite maturation. Geochim. Cosmochimica Acta, 61, 2921-2928.

Michels, R., Landais, P., Torkelson, B.E. & Philp, R.P. 1995b. Effects of effluents and water pressure on oil generation during confined pyrolysis and high-pressure hydrous pyrolysis. Geochimica et Cosmochimica Acta, 59, 1589-1604.

Uguna, C.N., Snape, C.E., Meredith, W., Carr, A.D., Scotchman, I.C. and Davis, R.C. in preparation. High-pressure liquid water pyrolysis of coal to investigate maturation and hydrocarbon generation in geological basins. Organic Geochemistry

Figure 1. Bitumen and gas yields generated from Kimmeridge Clay (type II), Longannet and Svalbard coals at 350^oC for 24 hours under non-hydrous, hydrous and 500 bar water pressure.