| Popis: |
The objective of this project was to develop a membrane process for the denitrogenation of natural gas. Large proven reserves in the Lower-48 states cannot be produced because of the presence of nitrogen. To exploit these reserves, cost-effective, simple technology able to reduce the nitrogen content of the gas to 4-5% is required. Technology applicable to treatment of small gas streams (below 10 MMscfd) is particularly needed. In this project membranes that selectively permeate methane and reject nitrogen in the gas were developed. Preliminary calculations show that a membrane with a methane/nitrogen selectivity of 3 to 5 is required to make the process economically viable. A number of polymer materials likely to have the required selectivities were evaluated as composite membranes. Polyacetylenes such as poly(1-trimethylsilyl-1-propyne) [PTMSP] and poly(4-methyl-2-pentyne) [PMP] had high selectivities and fluxes, but membranes prepared from these polymers were not stable, showing decreasing flux and selectivity during tests lasting only a few hours. Parel, a poly(propylene oxide allyl glycidyl ether) had a selectivity of 3 at ambient temperatures and 4 or more at temperatures of {minus}20 C. However, Parel is no longer commercially available, and we were unable to find an equivalent material in the time available. Therefore, most of our experimental work focused on silicone rubber membranes, which have a selectivity of 2.5 at ambient temperatures, increasing to 3-4 at low temperatures. Silicone rubber composite membranes were evaluated in bench-scale module tests and with commercial-scale, 4-inch-diameter modules in a small pilot plant. Over six days of continuous operation at a feed gas temperature of {minus}5 to {minus}10 C, the membrane maintained a methane/nitrogen selectivity of about 3.3. Based on the pilot plant performance data, an analysis of the economic potential of the process was prepared. We conclude that a stand-alone membrane process is the lowest-cost technology for small gas streams containing less than 10% nitrogen. The membrane process can recover more than 60-70% of the hydrocarbon content of the gas at a cost of $0.60-0.70/Mscfd. The capital cost of the process is about $100-200/Mscf. A number of small operators appear to be ready to use the technology if these costs can be demonstrated in the field. A second, and perhaps better, application of the technology is to combine the membrane process with a cryogenic process to treat large gas streams containing 10-20% nitrogen. The combination process achieves significant synergies. The membrane process performs a bulk separation of the gas, after which the cryogenic process treats the membrane residue (nitrogen-enriched) gas to recover more methane. Overall, hydrocarbon recoveries are greater than 95%. The capital cost of the combination process is lower than that of either process used alone and the processing costs are in the range $0.30-0.40/Mscf. This operating cost would be attractive to many gas producers. MTR is collaborating with a producer of cryogenic systems to further develop the combination process. A number of innovations in membrane process designs were made during the project; four U.S. patents covering various aspects of the technology were filed and issued. |
