Differential solubilities of CO[2] and methane and the potential for coupling in TFA

We wish to add to the recent insightful comments of Wilcox et al. [1] regarding the thermodynamics of light alkane carboxylation. This process is particularly of industrial interest when it comes to methane which is vented at oil production sites where there are significant quantities of "associated" natural gas also being released. This is of course damaging environmentally. There are also many natural gas reservoirs in locations where there is no gas markets within a pipe-able distance. Thus the methane is "stranded" and any conversion to a liquid, such as the carboxylation discussed by Wilcox et al. is interesting from the point of making gas into a transportable liquid. Finally, the possibility of simultaneously capturing CO2 as reported by Taniguchi et al. [2] is of great significance given the increasing occurrence of gas wells with >15% CO2 content which are not considered to be currently commercially exploitable. The thermodynamics of the coupling process for CO2 with methane is indeed extremely unfavorable. However, we would like to augment the comments in the recent letter by pointing out that were the reaction to be carried out under the conditions such as those found at the bottom of an oil well bore hole, then the pressure could be sufficient to cause dew-pointing of the product. The miniscule equilibrium constant need not necessarily rule out the reaction. Providing the coupled product can be removed from the gas phase then the reaction could proceed to 100% completion. Our calculations [3] show that in order to dew-point a mixture at a typical well bore pressure of 70 bar one would need about 20% conversion of methane and this is indeed way beyond the conversions calculated by the authors. However, what both the authors of the letter and the original article seem to have missed is the two phase nature of the reaction. Taniguchi et al. based their measurement on reactant inputs of the measured partial pressures of 5 and 20 bar of methane and CO2, respectively. The calculated number of moles of reactants which they quote—0.95 mmol of CH4 and 3.78 mmol of CO2—would appear to be based on these pressures, but these are not correct because considerable quantities of CO2 and to a lesser extent methane, dissolve in a polar solvent such as trifluoroacetic acid. This is to be expected from classic gas solubility interactions-indeed the standard technology for cleaning up gas streams involves selective dissolution and partial reaction of CO2 with a reactive solvent such as ethanolamine [4]. Of the total volume of the reactor in the CH4–CO2 coupling experiment (25 ml) 20 ml was filled with liquid leaving only 5 ml volume for a gas cap. The assumption seems to have been that the partial pressures of methane and CO2 in this gas cap defined the number of reacting moles of each species. However, this is not correct as can be seen from the solubility curves which we have calculated from an extended equation of state program based on a cubic equation of state of the Soave–Redlich–Kwong type with pure component parameters fitted to vapor pressures and liquid densities along with a composition dependent mixing rule. Fig. 1 shows the partial pressures of each of the components for the additions of methane and CO2 (each in the absence of the other). We note that at 5 bar partial pressure of methane, actually 3.2 mol of methane have been added—the original authors claimed 0.95 mmol. For 20 bar of CO2, 42 mol of that gas have been added compared to the 3.78 mmol claimed and which we believe to have been based on a trapped gas cap, although in fact the slope of the line indicates that above 12 bar-the effective dew pressure—we are adding CO2 in the liquid phase. (This can be confirmed by examining the partition coefficients given by the ratio of the component mole fractions in the gas phase to that in the liquid phase.) At any rate, under the reaction conditions, the true amount of reactant methane is 3.4 times that claimed and the amount of CO2 is 11 times.