Modeling Early Methane Generation in Coal

Abstract

Thermogenic methane yields can be estimated indirectly from the average elemental composition of coals of different rank or inferred from the results of coal pyrolysis experiments. Unfortunately, most published studies have been insufficiently detailed to estimate gas contents in lignite, subbituminous coal and high-volatile B bituminous coal. In addition, we note that the theoretical coalbed methane generation curves of Juntgen and Karweil and other commonly quoted papers overestimate methane yield because they do not consider hydrogen loss from coal in the form of water. In order to place better constraints on the economic potential of methane in lowrank coals, anhydrous, sealed-tube pyrolysis experiments were carried out on a Paleocene lignite from North Dakota. Experiments were conducted at heating rates of 10 °C/h and 10 °C/day between temperatures of 100 and 454 °C. With increasing final pyrolysis temperature, mean random huminite/vitrinite reflectance values increased from 0.31 to 1.61%, atomic H/C values of the extracted coal decreased from 0.88 to less than 0.56, and methane yields increased to a maximum of 46 mL/g initial lignite, or approximately 1560 cf/ton (dry, ash-free basis) (cf ) cubic feet). Based on these results, coalification to high-volatile A bituminous rank or higher (R o g 0.8%, atomic H/C e 0.72, and NAI [log(n-C 16 /n-C 30 )] g 0.03) appears required to achieve a modest in situ economic threshold of 300 cf/ton methane. Pyrolysis yields were used to model early methane generation with a series of parallel, first-order reactions with activation energies between 41 and 54 kcal/mol and a single frequency factor of 9.88 × 10 11 s -1 . Extrapolation of these parameters and a modified version of the EASY%Ro vitrinite reflectance model to geologic heating rates suggests that T > 120 °C and R o g 0.9% are required to exceed the 300 cf/ton threshold. We conclude that while methane concentrations greater than 300 cf/ton may be found in highvolatile B bituminous and lower rank coals, in most cases they must be attributed to migrated gas or to near-surface (e3000 ft) microbial activity.

The generation kinetics of natural gases in the Kela-2 gas field from the Kuqa Depression, Tarim Basin, northwestern China

“…The above variation trend of the methane carbon isotopic ratios is similar to that of coals and type-Ⅲ kerogens reported by Tang et al (1996Tang et al ( , 2000 and Lorant et al (1998) in the confined system. It can also be compared with the general characteristics described by Berner et al (1995) and Cramer et al (1998Cramer et al ( , 2001 in the open system.…”

“…From previous studies (Tang et al, 1996), we used a discrete activation energy model to fit the experimental data for two different heating rates.…”

“…The kinetic parameters for methane generation and carbon isotope fractionation based on the pyrolysis experiment data were calculated and fitted by the method described by Tang et al (1996Tang et al ( , 2000. Shown in Fig.…”

Hydrocarbon generation from coal, extracted coal and bitumen rich coal in confined pyrolysis experiments

“…Hydrocarbon generation from kerogen is generally described using a set of parallel first order reactions with a single frequency factor and a distribution of activation energies (e.g., Ungerer and Pelet, 1987;Schenk and Horsfield, 1993;Tang et al, 1996;Behar et al, 1997). For coal samples, simplified kinetic models with a single frequency factor are not sufficient to correctly predict hydrocarbon generation under geological conditions (e.g., Dieckmann, 2005).…”

Modeling of gas generation from the Cameo coal zone in the Piceance Basin, Colorado

“…However, the validity of using R o to indicate organic maturity and to interpret the thermal history in a sedimentary basin depends on how well the kinetics of vitrinite maturation is understood. Several published R o kinetic models that purport to be globally applicable exist; however, they consistently fail to perform in some areas of the world (Buiskool Toxopeus, 1983;Ungerer, 1990;Throndsen et al, 1993;Tang et al, 1996). This underscores the need for the development of basin-specific R o kinetic models.…”