Bench experiments and laboratory field pilot plant testing have led to the development of a solvent that is selective in removing CO and H S from natural gas at pipeline pressure. The acid gases are physically absorbed in the solvent and no chemical reaction takes place. Introduction A few years ago, most commercial acid-gas removal processes involved the use of either amines or processes involved the use of either amines or potassium carbonate solutions. These materials chemically potassium carbonate solutions. These materials chemically react with hydrogen sulfide and carbon dioxide, giving off appreciable quantities of heat. Heat must be supplied to regenerate the solutions. In recent years, several companies have developed selective solvents to remove acid gases from natural gas and synthesis gas. The selective solvents operate on the principle of pure physical absorption, as one would calculate using vapor-liquid equilibrium ratios (K-values) for the appropriate systems. Since there is a heat of absorption and desorption involved, the heat effects are much smaller than with solutions involving chemical reactions, and the selective solvent can be regenerated without reboiling. Solvent absorption is especially useful in cases where the partial pressure of the acid gas exceeds 150 psia, and preferably 200 psia. The cost advantage of the solvent absorption psia. The cost advantage of the solvent absorption process arises largely from the saving of heat process arises largely from the saving of heat ordinarily used in the regeneration of amines and hot potassium carbonate solutions, which is substantial potassium carbonate solutions, which is substantial when large volumes of acid gas are involved. Preliminary Screening Tests Preliminary Screening Tests Several years ago, we made a detailed study of acid gas processing that led to the discovery of some 11 classes of polar, organic solvents that were effective in removing acid gas. Solubilities of CO and propane were measured in a large number of solvents at 1 atm and about 80 deg. F. Results of these measurements revealed the solvents covered by Union Oil Co. patents. Solubility data related to these solvents and patents. Solubility data related to these solvents and competitive solvents are shown in Table 1. The solubility ratio of CO to propane is a measure of the selectivity of the solvent. A high selectivity is obviously desirable. Some H S solubility data have been included. Table 2 lists properties of these relatively nonvolatile selective solvents. Results from a bench scale pilot plant narrowed the number of solvents of chief interest, and pointed to the need for a field pilot plant for further study. In the fall of 1964, a pilot plant was installed near the Lisbon field, Moab, Utah, to test some of our solvents as well as a competitive solvent on a sour-gas stream; results of the pilot plant tests on methyl cyanoacetate (MCA) and propylene carbonate (PC) are reported here. The pilot plant is still in use to treat fuel gas for gas compressors. Description of Field Pilot Plant The Steams-Roger Corp. of Denver was selected to design and install the field pilot plant, which is shown in Fig. I in a simplified flow diagram. The pilot plant processed up to 1.8 MMcf/D at a pressure of 770 processed up to 1.8 MMcf/D at a pressure of 770 psig during the runs described here. Raw gas psig during the runs described here. Raw gas contained about 30 percent CO and 1 percent H S. Treated gas was obtained with as little as a few tenths of a percent of CO and less than 1/4 grain of H S per 100 scf of gas. The major equipment included in the pilot plant is described below. JPT P. 483