Integrated Ceramic Membrane System for Hydrogen Production

Schwartz, Joseph ; Lim, Hankwon ; Drnevich, Raymond

Praxair, Inc.

关键词: Removal;    Hydrogen Production;    Alloys;    Membranes;    08 Hydrogen;   

DOI  :  10.2172/984651
RP-ID  :  DOE/GO/10534-1
RP-ID  :  FC36-00GO10534
RP-ID  :  984651

美国|英语

来源: UNT Digital Library

【 摘 要 】

Phase I was a technoeconomic feasibility study that defined the process scheme for the integrated ceramicmembrane system for hydrogen production and determined the plan for Phase II. The hydrogenproduction system is comprised of an oxygen transport membrane (OTM) and a hydrogen transportmembrane (HTM).Two process options were evaluated: 1) Integrated OTM-HTM reactor – in this configuration, the HTMwas a ceramic proton conductor operating at temperatures up to 900°C, and 2) Sequential OTM and HTMreactors – in this configuration, the HTM was assumed to be a Pd alloy operating at less than 600°C. Theanalysis suggested that there are no technical issues related to either system that cannot be managed. Theprocess with the sequential reactors was found to be more efficient, less expensive, and more likely to becommercialized in a shorter time than the single reactor. Therefore, Phase II focused on the sequentialreactor system, specifically, the second stage, or the HTM portion. Work on the OTM portion wasconducted in a separate program.Phase IIA began in February 2003. Candidate substrate materials and alloys were identified and porousceramic tubes were produced and coated with Pd. Much effort was made to develop porous substrateswith reasonable pore sizes suitable for Pd alloy coating. The second generation of tubes showed someimprovement in pore size control, but this was not enough to get a viable membrane. Furtherimprovements were made to the porous ceramic tube manufacturing process. When a support tube wassuccessfully coated, the membrane was tested to determine the hydrogen flux. The results from all thesetests were used to update the technoeconomic analysis from Phase I to confirm that the sequentialmembrane reactor system can potentially be a low-cost hydrogen supply option when using an existingmembrane on a larger scale.Phase IIB began in October 2004 and focused on demonstrating an integrated HTM/water gas shift(WGS) reactor to increase CO conversion and produce more hydrogen than a standard water gas shiftreactor would. Substantial improvements in substrate and membrane performance were achieved inanother DOE project (DE-FC26-07NT43054). These improved membranes were used for testing in awater gas shift environment in this program. The amount of net H2 generated (defined as the difference ofhydrogen produced and fed) was greater than would be produced at equilibrium using conventional watergas shift reactors up to 75 psig because of the shift in equilibrium caused by continuous hydrogenremoval. However, methanation happened at higher pressures, 100 and 125 psig, and resulted in less netH2 generated than would be expected by equilibrium conversion alone. An effort to avoid methanation bytesting in more oxidizing conditions (by increasing CO2/CO ratio in a feed gas) was successful and net H2generated was higher (40-60%) than a conventional reactor at equilibrium at all pressures tested (up to125 psig). A model was developed to predict reactor performance in both cases with and withoutmethanation. The required membrane area depends on conditions, but the required membrane area isabout 10 ft2 to produce about 2000 scfh of hydrogen. The maximum amount of hydrogen that can beproduced in a membrane reactor decreased significantly due to methanation from about 2600 scfh toabout 2400 scfh. Therefore, it is critical to eliminate methanation to fully benefit from the use of amembrane in the reaction. Other modeling work showed that operating a membrane reactor at highertemperature provides an opportunity to make the reactor smaller and potentially provides a significantcapital cost savings compared to a shift reactor/PSA combination.