Adsorption of Pure Methane, Nitrogen, and Carbon Dioxide and Their Mixtures on San Juan Basin Coal

Abstract

The major objectives of this project were to (a) measure the adsorption behavior of pure methane, nitrogen, CO{sub 2} and their binary and ternary mixtures on wet Tiffany coal at 130 F and pressures to 2000 psia; (b) correlate the equilibrium adsorption isotherm data using the extended Langmuir model, the Langmuir model, the loading ratio correlation and the Zhou-Gasem-Robinson equation of state; and (c) establish sorption-time estimates for the pure components. Specific accomplishments are summarized below regarding the complementary tasks involving experimental work and data correlation. Representative coal samples from BP Amoco Tiffany Injection Wells No.1 and No.10 were prepared, as requested. The equilibrium moisture content and particle size distribution of each coal sample were determined. Compositional coal analyses for both samples were performed by Huffman Laboratories, Inc. Pure gas adsorption for methane on wet Tiffany coal samples from Injection Wells No.1 and No.10 was measured separately at 130 F (327.6 K) and pressures to 2000 psia (13.7 MPa). The average expected uncertainty in these data is about 3% (9 SCF/ton). Our measurements indicate that the adsorption isotherms of the two coal samples exhibit similar Langmuir-type behavior. For the samples from the two wells, a maximum variation of about 5% in the amount adsorbed is observed at 2000 psia. Gas adsorption isotherms were measured for pure methane, nitrogen and CO{sub 2} on a wet, mixed Tiffany coal sample. The coal sample was an equal-mass mixture of coals from Well No.1 and Well No.10. The adsorption measurements were conducted at 130 F at pressures to 2000 psia. The adsorption isotherms have average expected experimental uncertainties of 3% (9 SCF/ton), 6% (8 SCF/ton), and 7% (62 SCF/ton) for methane, nitrogen, and CO{sub 2}, respectively. Adsorption isotherms were measured for methane/nitrogen, methane/CO{sub 2} and nitrogen/CO{sub 2} binary mixtures on wet, mixed Tiffany coal at 130 F and pressures to 2000 psia. These measurements were conducted for a single molar feed composition for each mixture. The expected uncertainties in the amount adsorbed for these binary mixtures vary with pressure and composition. In general, average uncertainties are about 5% (19 SCF/ton) for the total adsorption; however, the expected uncertainties in the amount of individual-component adsorption are significantly higher for the less-adsorbed gas at lower molar feed concentrations (e.g., nitrogen in the 20/80 nitrogen/CO{sub 2} system). Adsorption isotherms were measured for a single methane/nitrogen/CO{sub 2} ternary mixture on wet, mixed Tiffany coal at 130 F and pressures to 2000 psia. The nominal molar feed composition was 10/40/50. The average expected uncertainty for the total adsorption and CO{sub 2} adsorption is about 5% (16 SCF/ton). However, the low adsorption of nitrogen and methane in this ternary yield average experimental uncertainties of 14% (9 SCF/ton) and 27% (9 SCF/ton), respectively. Limited binary and ternary gas-phase compressibility factor measurements at 130 F and pressures to 2000 psia involving methane, nitrogen, and CO{sub 2} were conducted to facilitate reduction of our ternary adsorption data. These newly acquired data (and available data from the literature) were used to improve the Benedict-Webb-Rubin (BWR) equation-of-state (EOS) compressibility factor predictions, which are used in material balance calculations for the adsorption measurements. In general, the optimized BWR EOS represents the experimental compressibility factor data within 0.5% AAD. The Langmuir/loading ratio correlation (LRC) and the Zhou-Gasem-Robinson (ZGR) two-dimensional EOS were used to analyze the newly acquired adsorption data. Model parameters were obtained for the systems studied. The LRC and ZGR EOS were used to correlate the adsorption data for methane, nitrogen, and CO{sub 2} and their mixtures on wet Tiffany coal. The model parameters were determined by minimizing the sum of squares of weighted errors in the calculated amounts of gas adsorbed. The results demonstrate the ability of the LRC and ZGR EOS to represent the total pure, binary and ternary systems within their expected experimental uncertainties. Specifically, representations with average absolute percentage errors (AAD) of 1-3% (2-15 SCF/ton), 1-8% (1-25 SCF/ton), and 2-10% (7-37 SCF/ton) were obtained for the pure, total binary, and total ternary adsorption isotherms, respectively. However, the quality of fit for the individual-component adsorption varies significantly, ranging from 3% for the more-adsorbed methane or CO{sub 2} to 32% for the less-adsorbed nitrogen. The LRC and ZGR EOS are capable of predicting binary adsorption isotherms based solely on pure-fluid adsorption parameters within twice their experimental uncertainties (1-50 %AAD, 5-40 SCF/ton).